The implementation of neural networks (NN) on edge devices enables local processing of wireless data but faces challenges such as high computational complexity and memory requirements when deep neural networks (DNN) are used. Shallow neural networks customized for specific problems are more efficient, requiring fewer resources, and resulting in a lower latency solution. An additional benefit of the smaller network size is that it is suitable for real-time processing on edge devices. The main concern with shallow neural networks is their accuracy performance compared to DNNs. In this paper, we demonstrate that a customized adaptive activation function (AAF) can meet the accuracy of a DNN. We designed an efficient FPGA implementation for a customized segmented spline curve neural network (SSCNN) structure to replace the traditional fixed activation function with an AAF. We compared our SSCNN with different neural network structures such as real-valued time delay neural network (RVTDNN), augmented real-valued time delay neural network (ARVTDNN), and deep neural networks with different parameters. Our proposed SSCNN implementation uses 40% fewer hardware resources and no Block RAMS compared to the DNN with similar accuracy. We experimentally validate this computationally efficient and memory-saving FPGA implementation of SSCNN for digital predistortion of RF power amplifiers using the AMD/Xilinx RFSoC ZCU111. The implemented solution uses less than 3% of the available resources. The solution also enables an increase of the clock frequency to 221.12 MHz, allowing the transmission of wide bandwidth signals.

Tight gas, a category of Deep Green Energy Resources, rely on advanced intelligent monitoring methods for their extraction. Conventional logging for reservoir evaluation relies on logging data and the manual setting of evaluation criteria to classify reservoirs. However, the complexity and heterogeneity of tight reservoirs pose challenges in accurately identifying target layers by using traditional well-logging techniques. Machine learning may hold the key to solving this problem, as it enables computers to learn without being explicitly programmed and manually added rules. Therefore, it is possible to make reservoir evaluation via using machine learning methods. In this paper, the reservoir quality index (RQI) and porous geometric parameters obtained from the optimized inversion of the spherical-tubular model are adopted to evaluate the reservoir. Then, three different machine learning approaches, the random forest (RF) algorithm, support vector machine (SVM) algorithm, and extreme gradient boosting (XGB) algorithm, are utilized for reservoir classification. The selected dataset covers more than 7,000 samples from five wells. The data from four wells are arranged as the training dataset, and the data of the remaining one well is designed as the testing dataset to calculate the prediction accuracies of different machine learning algorithms. Among them, accuracies of RF, SVM, and XGB are all higher than 90%, and XGB owns the highest result by reaching 97%. Machine learning based approaches can greatly assist reservoir prediction by implementing the well-logging data. The research highlights the application of reservoir classification with a higher prediction accuracy by combining machine learning algorithms with NMR logging based pore structure characterization, which can provide a guideline for sweet spot identification within the tight formation. This not only optimizes resource extraction but also aligns with the global shift towards clean and renewable energy sources, promoting sustainability and reducing the carbon footprint associated with conventional energy production. In summary, the fusion of machine learning and NMR logging based reservoir evaluation plays a crucial role in advancing both energy efficiency and the transition to cleaner energy sources.

Cancer metastasis accounts for approximately 90% of cancer deaths, and elucidating markers in metastasis is the first step in its prevention. To characterize metastasis marker genes of breast cancer (MGs), XGBoost models that classify metastasis status were trained with gene expression profiles from TCGA. Then, a metastasis score (MS) was assigned to each gene by calculating the inner product between the feature importance and AUC performance of the models. As a result, the 54, 202, and 357 genes with the highest MS were characterized as MGs by empirical P-value cutoffs of 0.001, 0.005, and 0.01, respectively. The three sets of MGs were compared with those from existing metastasis marker databases, which provided significant results in most comparisons. We noticed that the set of MGs with the median EP cutoff showed better performance than the other two sets, suggesting the importance of the cutoff used in determining MGs. They were also significantly enriched in biological processes associated to breast cancer metastasis. The MGs that could not be identified by statistical analysis (e.g., GOLM1, ELAVL1, UBP1, and AZGP1) as well as the MGs with the highest MS (e.g., ZNF676, FAM163B, LDOC2, IRF1, and STK40) were verified via the literature. Additionally, we checked how close the MGs are located to each other in the protein–protein interaction networks. We expect that the characterized markers will help understand and prevent breast cancer metastasis.

Deep learning is improving by leaps and bounds in remote sensing images (RSIs) analysis, pre-trained convolutional neural networks (CNNs) have shown remarkable performance in remote sensing image scene classification (RSISC). Nonetheless, pre-trained CNNs require massive annotated data as samples for data training. When labeled samples are not sufficient, the most common solution is to the pre-trained CNNs using a great deal of natural image dataset (e.g. ImageNet). However, these pre-trained CNNs require a large quantity of labelled data for training, which is often not feasible in RSISC, especially when the target RSIs have different imaging mechanisms from RGB natural images. In this paper, we proposed an improved hybrid classical-quantum transfer learning CNNs composed of classical and quantum elements to classify open-source RSI dataset. The classical part of the model is made up of a ResNet network which extracts useful features from RSI datasets. To further refine the network performance, a tensor quantum circuit is subsequently employed by tuning parameters on near-term quantum processors. We tested our models on open-source RSI dataset. In our comparative study, we have concluded that the hybrid classical-quantum transferring CNN has achieved better performance than other pre-trained CNNs based RSISC methods with small training samples. Moreover, it has been proved that the proposed algorithm improves the classification accuracy while greatly decreasing sum of model parameters and sum of training data.

Wearable sensors have the potential to enable comprehensive patient characterization and optimized clinical intervention. Critical to realizing this vision is accurate estimation of biomechanical time-series in daily-life, including joint, segment, and muscle kinetics and kinematics, from wearable sensor data. The use of physical models for estimation of these quantities often requires many wearable devices making practical implementation more difficult. However, regression techniques may provide a viable alternative by allowing the use of a reduced number of sensors for estimating biomechanical time-series. Herein, we review 46 articles that used regression algorithms to estimate joint, segment, and muscle kinematics and kinetics. We present a high-level comparison of the many different techniques identified and discuss the implications of our findings concerning practical implementation and further improving estimation accuracy. In particular, we found that several studies report the incorporation of domain knowledge often yielded superior performance. Further, most models were trained on small datasets in which case nonparametric regression often performed best. No models were open-sourced, and most were subject-specific and not validated on impaired populations. Future research should focus on developing open-source algorithms using complementary physics-based and machine learning techniques that are validated in clinically impaired populations. This approach may further improve estimation performance and reduce barriers to clinical adoption.

An active power dispatch method for a microgrid (MG) with multi-type loads, renewable energy sources (RESs) and distributed energy storage devices (DESDs) is the focus of this paper. The MG operates in a grid-connected model, and distributed power sources contribute to the service for load demands. The outputs of multiple DESDs are controlled to optimize the active power dispatch. Our goal with optimization is to reduce the economic cost under time-of-use (TOU) price, and to adjust the excessively high or low load rate of distributed transformers (DTs) caused by the peak-valley demand and load uncertainties. To simulate a practical environment, the stochastic characteristics of multi-type loads are formulated. The transition matrix of system state is provided. Then, a finite-horizon Markov decision process (FHMDP) model is established to describe the dispatch optimization problem. A learning-based technique is adopted to search the optimal joint control policy of multiple DESDs. Finally, simulation experiments are performed to validate the effectiveness of the proposed method, and the fuzzification analysis of the method is presented.

The use of drones for various applications has become increasingly popular in recent years, and machine learning has played a significant role in this trend. In this paper, we provide a comprehensive survey of the classification and application of machine learning in drones. The paper begins with an overview of the different types of machine learning algorithms and their applications in drones, including supervised learning, unsupervised learning, and reinforcement learning. Next, we present a detailed analysis of various real-world applications of machine learning in drones, such as object recognition, route planning, obstacle avoidance, search area optimization, and autonomous search. The paper also discusses the challenges and limitations of using machine learning in drones, such as data privacy, data quality, and computational requirements. Finally, the paper concludes with a discussion of the future directions of machine learning in drones and its potential impact on various industries and fields. This paper provides a valuable resource for researchers, practitioners, and students interested in the intersection of machine learning and drones.

Topic discovery is finding the main idea of large amounts of textual data. It indicates the recurring topics in the documents, allowing an overview of the texts. Current topic discovery models receive the texts, with or without pre-processing of Natural Language Processing. The processing consists of stopwords removal, text cleaning and normalization (lowercase conversion). A topic discovery model that receives texts with or without processing generates general topics since the input data is many uncategorized texts. The general topics do not offer a detailed overview of the input texts, and manual text categorization is a time-consuming and tedious task. Accordingly, it is necessary to integrate an automatic text classification task in the topic discovery process to obtain specific topics with their top words that contain relevant relationships based on belonging to a class. Text classification performs a word analysis that makes up a document to decide what class or category is being identified; then, integrating the text classification before a topic discovery process will provide latent topics depicted by top words with a high coherence in each topic based on the previously obtained classes. Therefore, this paper exposes a approach that integrates text classification into topic discovery from large amounts of English textual data, such as 20-Newsgroup and Reuters corpora. The text classification is accomplished with a Convolutional Neural Network(CNN) incorporating three embedding models based on semantic relationships. The topic discovery over categorized texts is realized with Latent Dirichlet Analysis(LDA), Probabilistic Latent Semantic Analysis(PLSA), and Latent Semantic Analysis(LSA) algorithms. An evaluation process was performed based on the normalized topic coherence metric. The 20-Newsgroup corpus was classified, and twenty topics with ten top words were discovered for each class, obtaining 0.1723 normalized topic coherence when applying LDA, 0.1622 with LSA, and 0.1716 with PLSA. The Reuters corpus was also classified, obtaining 0.1441 normalized topic coherence when applying the LDA algorithm to obtain 20 topics for each class.

ABSTRACT Background Ubiquitination plays an important role in protein post-translational processes and has been found to be involved in a number of regulatory functions including proteasome degradation, DNA repair, transcription, signal transduction, endocytosis, and sorting. As the identification of ubiquitination site is critical to furthering our understanding of the mechanism of ubiquitination, various experimental and machine learning methods have been used to conduct this task. It has been an important but challenging task to improve the accuracy of ubiquitination site prediction. In this research, we explore the possibility of improving the prediction performance of machine learning by incorporating grid search in the training process. Method We developed grid search procedures for each of six widely used machine learning methods including NB, LR, DT, SVM, LASSO, and KNN, and applied them to ubiquitination site prediction using the six PCP datasets that were previously developed. For each of the ML methods, we developed a set of values for each of the tunable hyperparameters available to the method. These sets of values then can be combined to form a large grid of hyperparameter settings, and each of these settings is used in the grid search. We integrated 5-fold cross-validation in grid search to train and test ML models and applied an additional independent validation procedure by conducting a pre-training 80-20 sample split. We evaluated the performance of the six methods by comparing them side by side for each of the six datasets. We also compared the grid search results with the results that were previously published without doing grid search. To optimize the prediction performance, we trained 1.1 million ML models in total through grid search. Results We compared the overall prediction performance of these six methods, as well as their prediction performance when working with balanced vs. imbalanced data, and large-scale vs. small-scale data. From the perspective of dataset, we find that the overall performance of every PCP dataset has been significantly improved in this study compared to the previous study, with the percentage increase of the average AUC of all six datasets ranging from 7.9% (PCP-4) up to 17.0% (PCP-1). From the perspective of method, we find that three out of four methods significantly benefit from grid search comparing to their previously published non-grid search results, with the maximum AUC improvement as high as 47% (LASSO on PCP-5), 43.3% (NB on PCP-1), and 33.7% (SVM on PCP-6). SVM overall ranks number one, followed by KNN as the number two performer based on their average AUCs on all datasets. But these two also ranked the top two (KNN 76 days and SVM 15 days) in terms of the total running time that they need to do grid search. We also find that SVM, KNN, and DT tend to handle small-scale and imbalanced datasets better, while LR, and LASSO are doing well with large-scale and balanced datasets. NB is more sensitive to data imbalance while less sensitive to the size of a dataset. Conclusions Our results show that using grid search has improved the performance of ubiquitination prediction significantly. We find that the performance of a method is closely related to its hyperparameter setting and the type of data it handles. Even though SVM is on average an outperformer, none of the methods can provide the best performance for all datasets. When sufficient computing resources are well accessible, grid search is an effective way to identify both a top performing model for a machine learning method and a suitable machine learning method for a particular dataset.

Classifying network flow subsequences in an Industrial IoT gateway is an effective method for identifying Industrial IoT faults. This paper proposes a subsequence classification algorithm SSGBUL-IKNN based on unsupervised learning encoding to address the problem of low classification accuracy. Firstly, we built a network module based on unsupervised learning to encode the flow sequence of the Industrial IoT gateway. In the encoding module, a subsequence calibration algorithm controls the increasing error during the coding process. Then, the specific fault type is obtained through an improved KNN classification algorithm based on the indication distance source from the integrated coding sequences converted by multi-dimensional network flow sequences. The test result on three Industrial IoT datasets of a sewage treatment plant shows that the accuracy of SSGBUL-IKNN in this paper exceeds 92%, significantly higher than the algorithm DTW and TSF. It reaches the classification requirements of Industrial IoT fault diagnosis.

In this report, we present a study demonstrating how machine learning methods can reveal the interaction of different risk factors of post-transplant leukemic relapse and obtain robust predictions even with a modest clinical dataset. Using a cohort of 63 pediatric patients with acute lymphocytic leukemia (ALL) and 46 patients with acute myeloid leukemia (AML) who underwent stem cell transplant at a single institution, we built predictive models of leukemic relapse with both pretransplant and posttransplant patient variables (specifically lineage-specific chimerism) using the random forest classifier. The random forest classifier revealed different important predictive factors between ALL and AML in our relapse models consistent with previous knowledge. Furthermore, it also distinguished donor CD34 chimerism as most impactful in relapse prediction compare to donor chimerism of other cellular lineages in our dataset for ALL but CD3 for AML. Our models greatly improved sensitivity and specificity at predicting relapses in cross validation compared to a reference model based on our own institutional incidence of relapse and inferential statistical principles. Local Interpretable Model-Agnostic Explanations, an interpretable machine learning tool, confirmed our Random Forest Classification result and provided an intuitive explanation of how our machine learning models made the relapse prediction for each individual patient.

When dealing with limited sample sizes or missing values in time series data, achieving high accuracy in forecasting becomes challenging. This challenge is particularly common in wind energy forecasting, where many wind farms lack complete data. In this paper, we employ the widely-used ARIMA (Autoregressive Integrated Moving Average) model for time series forecasting. To address the difficulties posed by limited sample sizes and missing values, we adopt a unique approach by utilizing multiple ARIMA models for different datasets. A key distinction from other multiple models methods for time series forecasting lies in each model within our approach using its own dataset, making the fusion process more intricate. To overcome this, we incorporate neural networks and two techniques for model fusion: direct fusion with a learning mechanism and compensation for missing data through training. For each ARIMA model, we introduce a noise estimation method to circumvent the white noise assumptions inherent in ARIMA models. Finally, we apply these methods to wind energy prediction, and our experimental results showcase a significant enhancement in forecasting accuracy.

Stress can reduce the accuracy of radiation therapy owing to muscle contraction and changes in breathing patterns. This study aimed to predict stress in patients before each treatment session using artificial intelligence (AI) from biological signals to enhance treatment accuracy. We measured 123 stress cases in 41 patients and calculated stress scores by analyzing seven stress-related features derived from heart rate variability obtained from photoplethysmography. The study analyzed stress score distribution and observed its trend changes throughout the treatment. Before-treatment information was used to predict changes in stress features during treatment. AI models included both non-pre-trained (Decision Tree, Random Forest, Support Vector Machine, Long Short-Term Memory (LSTM), Transformer) and pre-trained (ChatGPT) models. The performance was evaluated using 10-fold cross-validation, exact match ratio, accuracy, recall, precision, and F1 score. More than 90% of the patients experienced stress during radiation therapy. From all the classifications, LSTM and prompt engineering GPT4.0 had the highest accuracy (feature classification, LSTM: 0.703, GPT4.0: 0.659, stress classification: LSTM: 0.846, GPT4.0: 0.769). Our research pioneers the use of AI and biological signals for stress prediction in radiation therapy, potentially identifying patients needing psychological support, and suggesting methods to improve radiotherapy effectiveness through stress management.

Binarization of degraded documents is an important preprocessing task for various document analysis such as OCR and historical document analysis. Existing studies have applied various convolutional neural network (CNN) models and generative models for document binarization, but they do not show generalized performance for noise that the model has not seen and it suffers from extracting elaborate text strokes. In this paper, to overcome these challenges, we utilize latent diffusion model (LDM), which is known for high-quality image generation model, for the first time in document binarization. By utilizing the iterative diffusion-denoising process in latent space, it shows high-quality cleaned binarized image generation and high generalized performance through using both data distribution and time step while training. Additionally, we apply gated U-Net to the backbone network to preserve text strokes using trainable gating value. Gated convolution can extract elaborate text stroke by allowing the model to focus on text region by combining gating value and feature. Furthermore, we maximize the effectiveness of the proposed model by training it with a combination of LDM loss and pixel-level loss, which is suitable for the model structure. Experiments on H-DIBCO and DIBCO benchmark datasets show that the proposed model outperforms existing methods.

Federated learning techniques aim to train and build machine learning models based on distributed datasets across multiple devices, avoiding data leakage. The main idea is to perform training on remote devices or isolated data centers without transferring data to centralized repositories, thus mitigating privacy risks. Data analytics in education, in particular learning analytics, is a promising scenario to apply this approach to address the legal and ethical issues related to processing sensitive data. Indeed, given the nature of the data to be studied (personal data, educational outcomes, data concerning minors), it is essential to ensure that the conduct of these studies and the publication of the results provide the necessary guarantees to protect the privacy of the individuals involved and the protection of their data. In addition, the application of quantitative techniques based on the exploitation of data on the use of educational platforms, student performance, use of devices, etc., can account for educational problems such as the determination of user profiles, personalized learning trajectories, or early dropout indicators and alerts, among others. This paper presents the application of federated learning techniques to two learning analytics problems: dropout prediction and unsupervised student classification. The experiments allow us to conclude that the proposed solutions achieve comparable results from the performance point of view with the centralized versions, avoiding centralizing the data for training the models.

We study the brain segmentation by dividing the brain into multiple tissues. Given possible brain segmentation by deep, machine learning can be efficiently exploited to expedite the segmentation process in the clinical practice. To accomplish segmentation process, a MRI and tissues transfer using generative adversarial networks is proposed. Given the better result, we propose the transfer model using GAN. For the case of the brain tissues, white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) are segmented. Empirical results show that this proposed model significantly improved segmentation results compared to the stat-of-the-art results. Furthermore, a dice coefficient (DC) metric is used to evaluate the model performance.

This article is a theoretical contribution to mental model research, which currently has different threads. Whereas some researchers focus on the perceived causal structure, others also include decision policies and decisions. We focus on the link between recognized causal structure (“mental models of dynamic systems”) and policies, proposing Johnson-Laird’s theory of mental models as the link. The resulting framework hypothesizes two types of systematic mental model errors: (1) misrepresentation of the system’s structure and (2) failure to deploy relevant mental models of possibilities. Examination of three experiments through this lens reveals errors of both types. Therefore, we propose that the cognitive theory of mental models opens a path to better understand how people construct their decision policies and develop interventions to reduce such mental model errors. The article closes by raising several questions for empirical studies of the reasoning process leading from mental models of dynamic systems to decision policies.

In recent years, online learning has become one of the most popular methods of educational delivery due to advances in technology, which has been made even more evident in the COVID-19 lockdown period. Online education has evolved into a distinct field of study within the educational system over the last few years. It is also important to note that parallel with the growth in this field, there has also been an increase in the number of scholarly journals that regularly publish research in this field, reflecting the importance of this field in the modern day. In spite of the fact that online learning offers a wide range of educational options, from short courses to full-time degrees, as well as being accessible, flexible, environmentally friendly, and affordable, there are also certain challenges associated with this educational approach. These challenges include the lack of social interaction, technical errors, a lack of hands-on training, and difficulties in assessing students. It is, therefore, imperative to ask the crucial question of whether online learning can replace traditional classroom learning or whether it can supplement it in hybrid models with it, as well as what factors and conditions are likely to determine this in the short- and long-term, as well as how it will be blended together in the future. The purpose of this commentary is to provide a brief summary of the current status of both learning models, as well as their pros and cons, in order to answer the question that was posed above.

The popularity of deep reinforcement learning (DRL) methods in economics have been exponentially increased. DRL through a wide range of capabilities from reinforcement learning (RL) and deep learning (DL) for handling sophisticated dynamic business environments offers vast opportunities. DRL is characterized by scalability with the potential to be applied to high-dimensional problems in conjunction with noisy and nonlinear patterns of economic data. In this work, we first consider a brief review of DL, RL, and deep RL methods in diverse applications in economics providing an in-depth insight into the state of the art. Furthermore, the architecture of DRL applied to economic applications is investigated in order to highlight the complexity, robustness, accuracy, performance, computational tasks, risk constraints, and profitability. The survey results indicate that DRL can provide better performance and higher accuracy as compared to the traditional algorithms while facing real economic problems at the presence of risk parameters and the ever-increasing uncertainties.

Lettuce (Lactuca sativa) is a leafy vegetable that provides a valuable source of phytonutrients for a healthy human diet. Assessment of plant growth and composition is vital for determining crop yield and overall quality, however, classical laboratory analyses are slow and costly. Therefore, new, less expensive, more rapid, and non-destructive approaches are being developed, including those based on (hyper)spectral reflectance. Additionally, it is also important to determine how plant phenotypes respond to fertilizer treatments and whether these differences in response can be detected from analyses of hyperspectral image data. In the current study, we demonstrate the suitability of hyperspectral imaging in combination with machine learning models to estimate the content of chlorophyll (SPAD), anthocyanins (ACI), glucose, fructose, sucrose, vitamin C, β-carotene, N, P, K, dry matter content, and plant fresh weight. The implemented five classification and regression machine learning models showed high accuracy in classifying the lettuces by the applied fertilizers treatments and estimating nutrient concentrations. To reduce the input (predictor data, i.e., hyperspectral data) dimension, 13 principal components were found and applied in models. The implemented artificial neural network models of the machine learning algorithm demonstrated high accuracy (r = 0.85 ... 0.99) in estimating fresh leaf weight, and contents of chlorophyll, anthocyanins, N, P, K, and β-carotene. The four applied classification models of machine learning demonstrated 100% accuracy in classifying the studied baby leaf lettuces by phenotype when certain fertilizer treatments were applied.

Multimedia resources, such as instructional videos, are currently enjoying a certain popularity in training programmes for medical and dental students. The major challenge is to create such resources that have quality content and are acceptable to students. We investigated the opinions of a sample of 551 students from 4 medical universities in Romania in order to identify the characteristics of the instructional videos that students prefer and we comparatively analysed these characteristics according to gender, age group, but also according to their general preferences for using Internet services. We used univariate (hypothesis testing) and multivariate (Two-Step Clustering) data analysis techniques and revealed three clusters of students, primarily determined by perceptions of the visual appearance of instructional videos. The structure of the clusters by gender and age group is relatively similar, but we recorded differences between them associated with students' expressed preferences for certain Internet services compared to others. Students who prefer instructional videos to contain images used only for aesthetic purposes, to fill in breaks, use Internet services mainly for communication, while students who prefer in videos the use of already drawn explanatory diagrams are slightly younger than the others and use Internet services mainly for information and communication.

Lamb wave-based structural health monitoring is widely acknowledged as a reliable 11 method for damage identification, classification, localization and quantification. However, due to 12 the complexity of Lamb wave signals, especially after interacting with structural components and 13 defects, interpreting these waves and extracting useful information about the structure's health is 14 still a major challenge. However, deep learning-based strategy offers a great opportunity to address 15 such challenges as the algorithm can operate directly on raw discrete time-domain signals. Unlike 16 traditional methods, which often require careful feature engineering and preprocessing, deep learn-17 ing can automatically extract relevant features from the raw data. This paper proposes an autoen-18 coder based on a bidirectional long short-term memory network with maximal overlap discrete 19 wavelet transform layer to detect the signal anomaly and determine the location of the damage in 20 the composite structure. This approach has the potential to greatly enhance our ability to detect and 21 locate structural damage in composite structures, thereby increasing safety and efficiency.

To ensure the safety in aircraft flying, we aim use of the deep learning methods of nondestructive examination with multiple defect detection paradigms for X-ray image detection posed. The use of the Fast Region-based Convolutional Neural Networks (Fast R-CNN) driven model seeks to augment and improve existing automated Non-Destructive Testing (NDT) diagnosis. Within the context of X-ray screening, limited numbers insufficient types of X-ray aeronautics engine defect data samples can thus pose another problem in training model tackling multiple detections perform accuracy. To overcome this issue, we employ a deep learning paradigm of transfer learning tackling both single and multiple detection. Overall the achieve result get more then 90% accuracy based on the AE-RTISNet retrained with 8 types of defect detection. Caffe structure software to make networks tracking detection over multiples Fast R-CNN. We consider the AE-RTISNet provide best results to the more traditional multiple Fast R-CNN approaches simpler translate to C++ code and installed in the Jetson™ TX2 embedded computer. With the use of LMDB format, all images using input images of size 640 × 480 pixel. The results scope achieves 0.9 mean average precision (mAP) on 8 types of material defect classifiers problem and requires approximately 100 microseconds.

Diabetic retinopathy (DR) is the primary factor leading to vision impairment and blindness in diabetic people. Uncontrolled diabetes can damage the retinal blood vessels. Initial detection and prompt medical intervention are vital in preventing progressive vision impairment. Today's growing medical field poses more significant workload and diagnostic demands on medical professionals. In the proposed study, a convolutional neural network (CNN) is employed to detect the stages of DR. This research is crucial for studying DR because of its innovative methodology incorporating two different public datasets. This strategy enhances the model's capacity to generalize unseen DR images, as each dataset encompasses unique demographics and clinical circumstances. The network can learn and capture complicated hierarchical image features with asymmetric weights. Each image undergoes preprocessing using contrast-limited adaptive histogram equalization (CLAHE) and the discrete wavelet transform (DWT). The model is trained and validated using the combined datasets of Dataset for Diabetic Retinopathy (DDR) and the Asia Pacific Tele-Ophthalmology Society (APTOS). The CNN model is tuned with different learning rates and optimizers. The CNN model with Adam optimizer achieved 70% accuracy and an area under curve (AUC) score of 0.82. The recommended study results may reduce diabetes-related vision impairment by early DR severity identification.

Domain generalization has led to remarkable achievements in Hyperspectral Image (HSI) classification. Inspired by contrastive language-image pre-training (CLIP), the language-aware domain generalization method has been explored to learn cross-domain-invariant representation. However, existing methods face some challenges: 1) The weak capacity to extract long-range contextual information and inter-class correlation. 2) Due to the inadequacies of the large-scale pre-training for HSI data, the spatial-spectral features of HSI and linguistic features can not be straightforwardly alignment. To address the above problems, a novel network has been proposed with a CLIP framework, which consists of an image encoder, based on an encoder-only transformer to obtain the global contextual information and inter-class correlation, a frozen text encoder, and a cross-attention mechanism, named Linguistic-Interact-with-Visual Engager (LIVE), enhances the interaction between two modalities. Extensive experiments demonstrating superior performance over state-of-the-art methods in HSI Domain Generalization with a CLIP framework.

Detection of the region of interest (ROI) is a critical step in laryngeal image analysis for the delineation of glottis contour. The process can improve both computational efficiency and accuracy of the image segmentation task, which will facilitate subsequent analysis and characterization of the vocal fold vibration as it correlates with voice quality and pathology. This study aims to develop machine learning based approaches for automatic detection of ROI for glottis image sequences captured by high-speed video-endoscopy (HSV), a clinical laryngeal imaging modality. In particular, we first applied the supporting vector machine (SVM) method using histogram of oriented gradients (HOG) feature descriptor, and second, trained a convolutional neural network (CNN) model for this task. Comparisons are made for both approaches in terms of accuracy of recognition and computation time.

This research not only dedicated a less restrictive method of iteration-varying function for a learning control law to design a controller but also synchronize two nonlinear systems with free time-delay. In addition, the mathematical theory of system synchronization has proved rigorously and the theory verified through an example to demonstrate the behavior of each parameter in the theory. The design of a controller using the iterative learning control law is significant for robotic tracking. The controller in this research generates a feed-forward control input using the error dynamics among the drive-response systems. The error dynamics satisfies the Lyapunov function and the combination of output errors, which respectively represented relative estimated differences of the drive-response systems. The iterative learning control rule serves the function of a filter adding previous control error after the end of each iteration. The numerical example of a synchronous system is given a Lorenz system for driving and another with the iterative learning control law for response under different initial condition. The results verify and demonstrate the proposed mathematical theory. The simulation exhibits consistency in the behavior of each parameter to match mathematical theory.

This study leverages transformer-based models and focuses on data representation strategies in the named entity recognition task, including "single" (one sentence per vector), "merged" (multiple sentences per vector), and "context" (sentences joined with attention to context). Performance analysis reveals that models trained with a single strategy may not perform well on different data representations. A combined training procedure is proposed to address this limitation, using all three strategies to enhance the stability and adaptability of the model. The results of this approach are presented and discussed for various datasets for four languages (English, Polish, Czech, and German), demonstrating the effectiveness of the combined strategy.

The relationship between the street environment and the health, education, mobility, and criminal behaviors of its citizens has long been investigated by economists, sociologists and urban planners. Home buyers were found to pay a premium for better street appearance. Prior studies considering streetscapes mainly focus on objective measures such as the number of nearby trees, the tree canopy area, or the view index of physical features such as greenery, sky or building. However, subjective perceptions may have complex or subtle relationships to physical features, individual physical features or simply summing them up do not capture people’s comprehensive perception. In contrast, this study proposed a new approach for the urban-scale application to quantify both subjectively and objectively measured streetscape scores for six important perception qualities, namely Greenness, Walkability, Safety, Imageability, Enclosure, and Complexity. Built on prior quantitative studies in urban design quality and emerging applications in deep learning and open source street view imagery for urban perceptions, we integrated existing frameworks to (1) effectively collect and evaluate both subjectively and objectively- measured perceptions; (2) investigate the coherence and divergence in ML-predicted subjective scores and formula-derived objective scores; and (3) compare their effects in affecting house prices taking Shanghai as a case study using a large-scale dataset on home transactions. The results implied: first, the percentage increase in sales price attributable to street scores is significant for both subjective and objective measurements. In general subjective scores explained more variance over structural attributes and objective scores in hedonic price model. Particularly, objective Greenness, subjective Safety and Imageability scores positively affected house prices. Second, for Greenness and Imageability scores, the subjective and objective measures exhibited opposite signs in affecting house prices, which implied that there might be mechanisms related to the psychological, social-demographical characteristics of street users that have not been fully incorporated by objective measures that taking view indices or recombination of them. In addition, certain objective measure might outperform subjective counterpart when the connotation of the perception is self-evident and not complicated, for example the Greenness. For those concepts were not familiar to the average person, subjective framework exhibits better performance. This is the first study comprehensively expanding hedonic price method with both subjectively and objectively measured streetscape qualities. It suggested that city authorities could levy a street environment tax to compensate the public budget invested in street environment where developers secured benefits from a price premium. This study enriches our understanding of the economic values of the subjective and objective measures street qualities. It sheds light on promising future study areas where the coherence and divergence of the two measurements should be further stressed.

Climate change and development have increased urban flooding, requiring modernization of stormwater infrastructure. Retrofitting standard passive systems with controllable valves/pumps is promising, but requires real-time control (RTC). One method of automating RTC is reinforcement learning (RL), a general technique for sequential optimization and control in uncertain environments. The notion is that an RL algorithm can use inputs of real-time flood data and rainfall forecasts to learn a policy for controlling the stormwater infrastructure to minimize measures of flooding. In real-world conditions, rainfall forecasts and other state information, are subject to noise and uncertainty. To account for these characteristics of the problem data, we implemented Deep Deterministic Policy Gradient (DDPG), an RL algorithm that is distinguished by its capability to handle noise in the input data. DDPG implementations were trained and tested against a passive flood control policy. Three primary cases were studied: (i) perfect data, (ii) imperfect rainfall forecasts, and (iii) imperfect water level and forecast data. Rainfall episodes (100) that caused flooding in the passive system were selected from 10 years of observations in Norfolk, Virginia, USA; 85 randomly selected episodes were used for training and the remaining 15 unseen episodes served as test cases. Compared to the passive system, all RL implementations reduced flooding volume by 70.5% on average, and performed within a range of 5%. This suggests that DDPG is robust to noisy input data, which is essential knowledge to advance the real-world applicability of RL for stormwater RTC.

Electricity demand prediction is vital for energy production management and proper exploitation of the present resources. Recently, several novel machine learning (ML) models have been employed for electricity demand prediction to estimate the future prospects of the energy requirements. The main objective of this study is to review the various ML models applied for electricity demand prediction. Through a novel search and taxonomy, the most relevant original research articles in the field are identified and further classified according to the ML modeling technique, perdition type, and the application area. A comprehensive review of the literature identifies the major ML models, their applications and a discussion on the evaluation of their performance. This paper further makes a discussion on the trend and the performance of the ML models. As the result, this research reports an outstanding rise in the accuracy, robustness, precision and the generalization ability of the prediction models using the hybrid and ensemble ML algorithms.

Unmanned aerial vehicles (UAVs)-based communication system is a promising solution to meet coverage and capacity requirements of future wireless networks. However, UAV-enabled communications is constrained with its coverage, energy consumption, and flying regulations, and the number of works focusing on the sustainability aspect of UAV-assisted networking has been limited in the literature so far. In this paper, we propose a solution to this limitation; particularly, we design a $Q$-learning-based UAV positioning scheme for sustainable wireless connectivity considering key constraints, that are, altitude regulations, non-flight zones, and transmit power. The objective is to find the optimal position of the UAV base station (BS) and minimize the energy consumption while maximizing the number of users covered. Moreover, a weighting mechanism is developed, where the energy consumption and number of users covered can be prioritized according to network/battery conditions. The proposed Q-learning-based solution is compared to the baseline k-means clustering method, where the UAV BS is positioned at the centroid location that minimizes the cumulative distance between the UAV BS and the users. The results demonstrate that the proposed solution outperforms the baseline k-means clustering-based method in terms of the number of users covered while achieving the desired minimization of the energy consumption.

Substances that can modify the androgen receptor pathway in humans and animals are entering the environment and food chain with the proven ability to disrupt hormonal systems and leading to toxicity and adverse effects on reproduction, brain development, and prostate cancer, among others. State-of-the-art databases with experimental data of human, chimp, and rat effects by chemicals have been used to build machine learning classifiers and regressors and evaluate these on independent sets. Different featurizations, algorithms, and protein structures lead to dif- ferent results, with deep neural networks (DNNs) on user-defined physicochemically-relevant features developed for this work outperforming graph convolutional, random forest, and large featurizations. The results show that these user-provided structure-, ligand-, and statistically-based features and specific DNNs provided the best results as determined by AUC (0.87), MCC (0.47), and other metrics and by their interpretability and chemical meaning of the descriptors/features. In addition, the same features in the DNN method performed better than in a multivariate logistic model: validation MCC = 0.468 and training MCC = 0.868 for the present work compared to evalu- ation set MCC = 0.2036 and training set MCC = 0.5364 for the multivariate logistic regression on the full, unbalanced set. Techniques of this type may improve AR and toxicity description and predic- tion, improving assessment and design of compounds. Source code and data are available at https://github.com/AlfonsoTGarcia-Sosa/ML

The calculation of the global shear capacity of steel-concrete composite downstand cellular beams with precast hollow-core units is important as it affects the span to depth ratios and the amount of material used, hence affects the embodied CO2 calculation when designers are deciding on the floor grids. This paper presents a reliable tool that can be used by designers to alter and optimise grip options during the preliminary design stages, without the need to run onerous calculations. The global shear capacity prediction formula is developed using five machine learning models. First, a finite element model database is developed. The influence of the opening diameter, web opening spacing, tee-section height, concrete topping thickness, the interaction degree, and the number of shear studs above the web opening are investigated. Reliability analysis is conducted to assess the design method and propose new partial safety factors. The Catboost Regressor algorithm presented better accuracy compared to the other algorithms. An equation to predict the shear capacity of composite cellular beams with hollow-core units is proposed by Gene Expression Programming. In general, the partial safety factor for resistance, according to the reliability analysis, varied between 1.25 and 1.26.

In the development of a digital twin (DT) for hydropower turbines, dynamic modeling of the system (e.g., penstock, turbine, speed control) is crucial, along with all the necessary data interface, virtualization, and dashboard designs. Since the DT must mimic the actual dynamics of the hydropower turbine accurately, adaptive learning is required to train these dynamic models online so that the models in the DT can effectively follow the representation of the actual hydropower turbine dynamics accurately and reliably. This study presents an adaptive learning method for obtaining the hydropower turbine models for DT development of hydropower systems using recursive least squares algorithm. To simplify the formulation, the hydropower turbine under consideration was assumed to operate near a fixed operating point, where the system dynamics can be well represented by a set of linear differential equations with constant parameters. In this context, the well-known six-coefficient model for the Francis turbine was formulated as the starting point to obtain input and output models for the turbine. Then, an adaptive learning mechanism was developed to learn model parameters using the real-time data from a hydropower turbine testing system. This led to a semi-physical modeling in which the first principles and data-driven modeling are integrated to produce dynamic models for the DT development. Applications to a pilot system at the Norwegian University of Science and Technology (NTNU) were made, and the models learned adaptively using the data collected from the university’s pilot system. Desired modeling and validation results were obtained.

Learning disorders (LD) are diagnosed in children whose academic skills of reading, writing or mathematics are impaired and lagged according to their age, schooling and intelligence. Children with LD experience substantial working memory (WM) deficits, even more pronounced if more than one of the academic skills is affected. We compared the task-related EEG power spectral density of children with LD (n= 23), with a control group of children with good academic achievement (n= 22), during the performance of a WM task. sLoreta was used to estimate the current distribution at the sources, and 18 brain regions of interests (ROIs) were chosen with an extended version of the eigenvector centrality mapping technique. In this way, we lessen some drawbacks of the traditional EEG at the sensor space by an analysis at the brain sources level over data-driven selected ROIs. Results: The LD group showed fewer correct responses at the WM task, an overall slower EEG with more theta activity in all ROIs, less upper-alpha power at posterior areas, and less high-frequency beta activity in frontal areas. We explain these EEG patterns in LD children as indices of an inefficient neural resource management related with a delay in the neural development.

Optical networks are prone to power jamming attacks intending service disruption. This paper presents a Machine Learning (ML) framework for detection and prevention of jamming attacks in optical networks. We evaluate various ML classifiers for detecting out-of-band jamming attacks with varying intensities. Numerical results show that artificial neural network is the fastest ($10^6$ detection per second) for inference and most accurate ($\approx 100 \%$) in detecting power jamming attacks as well as identifying the optical channels attacked. We also discuss and study a novel prevention mechanism when the system is under active jamming attacks. For this scenario, we propose a novel resource reallocation scheme that utilizes the statistical information of attack detection accuracy to lower the probability of successful jamming of lightpaths while minimizing lightpaths' reallocations. Simulation results show that the likelihood of jamming a lightpath reduces with increasing detection accuracy, and localization reduces the number of reallocations required.

In recent years, weeds is responsible for most of the agricultural yield losses. To deal with this threat, farmers resort to spraying pesticides throughout the field. Such method not only requires huge quantities of herbicides but impact environment and humans health. One way to reduce the cost and environmental impact is to allocate the right doses of herbicide at the right place and at the right time (Precision Agriculture). Nowadays, Unmanned Aerial Vehicle (UAV) is becoming an interesting acquisition system for weeds localization and management due to its ability to obtain the images of the entire agricultural field with a very high spatial resolution and at low cost. Despite the important advances in UAV acquisition systems, automatic weeds detection remains a challenging problem because of its strong similarity with the crops. Recently Deep Learning approach has shown impressive results in different complex classification problem. However, this approach needs a certain amount of training data but, creating large agricultural datasets with pixel-level annotations by expert is an extremely time consuming task. In this paper, we propose a novel fully automatic learning method using Convolutional Neuronal Networks (CNNs) with unsupervised training dataset collection for weeds detection from UAV images. The proposed method consists in three main phases. First we automatically detect the crop lines and using them to identify the interline weeds. In the second phase, interline weeds are used to constitute the training dataset. Finally, we performed CNNs on this dataset to build a model able to detect the crop and weeds in the images. The results obtained are comparable to the traditional supervised training data labeling. The accuracy gaps are 1.5% in the spinach field and 6% in the bean field.

Self-supervised learning is a method that learns general representation from unlabeled data. Masked image modeling (MIM), one of the generative self-supervised learning methods, has drawn attention showing state-of-the-art performance on various downstream tasks, though showing poor linear separability resulting from the token-level approach. In this paper, we propose a contrastive learning-based multi-view masked autoencoder for MIM, exploiting an image-level approach by learning common features from two different augmented views. We strengthen MIM by learning long-range global patterns from contrastive loss. Our framework adopts simple encoder-decoder architecture, learning rich and general representation by following a simple process: 1) two different views are generated from an input image with random masking and by contrastive loss, we can learn semantic distance of the representations generated by an encoder. By applying a high mask ratio, 80%, it works as strong augmentation and alleviates the representation collapse problem. 2) With reconstruction loss, decoder learns to reconstruct an original image from the masked image. We assess our framework by several experiments on benchmark datasets of image classification, object detection, and semantic segmentation. We achieve 84.3% fine-tuning accuracy on ImageNet-1K classification and 76.7% in linear probing, exceeding previous studies and show promising results on other downstream tasks. Experimental results demonstrate that our work can learn rich and general image representation by applying contrastive loss to masked image modeling.

Inland waters pose a unique challenge for water quality monitoring by remote sensing techniques due to their complicated spectral features and small-scale variability. At the same time, collecting the high-quality reference data needed to calibrate remote sensing data products is both time consuming and expensive. In this study, we present the further development of a robotic team composed of an uncrewed surface vessel (USV) providing in situ reference measurements and an unmanned aerial vehicle (UAV) equipped with a hyperspectral imager. Together, this team is able to address the limitations of existing approaches by enabling the simultaneous collection of comprehensive in situ data together with high-resolution hyperspectral imagery. We showcase the capabilities of this team using data collected in a north Texas pond during three collections in the fall of 2020. Machine learning models for 13 variables are trained using the data set of paired in situ measurements and coincident reflectance spectra. These models can successfully estimate physical variables, including temperature, conductivity, pH, and turbidity, as well as the concentrations of blue-green algae, colored dissolved organic matter (CDOM), chlorophyll-a, crude oil, optical brighteners and the ions Ca2+, Cl−, and Na+. We extend the training procedure to utilize conformal prediction to estimate 90% confidence intervals for the output of each trained model. The maps generated by applying each trained model to the collected images reveal the small-scale spatial variability within the pond. Additionally, the permutation importance computed for each feature of the trained models highlights the spectral features relevant to each water quality variable.

Continual learning, a crucial facet of machine learning, involves the perpetual acquisition of valuable insights from incoming data, sans the necessity for full dataset access. Esteemed as a fundamental goal in artificial intelligence, continual learning grapples with an enduring challenge—catastrophic forgetting. A proficient model must exhibit adaptability for new data assimilation and robustness to retain existing knowledge. Class-incremental learning (CIL) aids the gradual integration of knowledge from newly introduced classes, forming a universal classifier. However, directly training the model with fresh class instances triggers a problem—forgetting distinguishing features of prior classes, causing a performance decline. Addressing such issues in machine learning, this survey aims to delineate significant challenges, outcomes, and recent advancements, including our contributions to CIL techniques, especially in image classification.

Sparse reward long horizon task is a major challenge for deep reinforcement learning algorithm. One of the key barriers is data-inefficiency. Even in the simulation environment, it usually takes weeks to training the agent. In this study, a data-efficiency training framework is proposed, where a curriculum learning is design for the agent in the simulation scenario. Different distributions of the initial state are set for the agent to get more informative reward during the whole training process. A fine-tuning of the parameters in the output layer of the neural network for value function is conduct to bridge the gap between sim-to-real. An experiment of UAV maneuver control is conducted in the proposed training framework to verify the method more efficient. We demonstrate that data-efficiency is different for the same data in different training stages.

Drones are being used more and more frequently and in various ways to attack people and critical infrastructures, which increases their level of danger and makes early detection and risk assessment even more difficult. There is no ultimate drone sensor. Rather, drones can only be effectively detected, if suitable sensor technologies are integrated within a system. Among the best suited sensorics for this purpose are radar and optical sensors, because the advantage of the radar, given by the long detection range, can be well combined with the additional visual-based information gained from a camera in the mid- and short-range, where usually the radar fails detecting micro drones. In this paper, a modular drone detection and assistance system MODEAS is presented, which can be used to automatically detect and track drones within a range up to some kilometers on the sky, but also in complex urban environments, where buildings and vegetation offer a perfect background for camouflage. The system provides an early detection of flying objects as potential drones, when those are not at all or only as small dots visible for the human eye. Main strengths of the system are the robust AI detection to differentiate such an object against clutter (birds, clouds) and its further classification by model and payload recognition, the latter aspect being considered as an indication of dangerousness. Moreover, once a drone is detected, the system follows it automatically, precisely and smoothly with a high-resolution pan-tilt-camera, measuring the distance to the sensor station with a laser rangefinder, compute the geocoordinates and visualize them live on a map. To improve the precision of the detection, a multisensorial fusion of all sensor detections is performed. Furthermore, alarms may be generated when a drone enters a restricted area.

This paper presents the development of a novel anti-violence device titled "no pAIn" (an acronym for Never Oppressed Protected by Artificial Intelligence Nonviolence system), which harnesses the power of artificial intelligence (AI). Primarily designed to combat violence against women, the device offers personal safety benefits for individuals across diverse demographics. Operating autonomously, it necessitates no user interaction post-activation. The AI engine conducts real-time speech recognition and effectively discerns genuine instances of aggression from non-violent disputes or conversations. Facilitated by its Internet connectivity, in the event of detected aggression, the device promptly issues assistance requests with real-time precise geolocation tracking to predetermined recipients, for immediate assistance. Its compact size enables discreet concealment within commonplace items like candy wrappers, or within purpose-built casings, or as wearable accessories. The device is battery-operated. The prototype was developed using a microcontroller board (Arduino Nano RP2040 Connect), incorporating an omnidirectional microphone and Wi-Fi module, all at a remarkably low cost. Subsequent functionality testing, performed in debug mode using the Arduino IDE serial monitor, yielded successful results. The AI engine exhibited exceptional accuracy in word recognition, complemented by a robust logic implementation, rendering the device highly reliable in discerning genuine instances of aggression from non-violent scenarios

The monitoring of surface-water quality followed by water-quality modeling and analysis is essential for generating effective strategies in water-resource management. However, worldwide, particularly in developing countries, water-quality studies are limited due to the lack of a complete and reliable dataset of surface-water-quality variables. In this context, several statistical and machine-learning models were assessed for imputing water-quality data at six monitoring stations located in the Santa Lucía Chico river (Uruguay), a mixed lotic and lentic river system. The challenge of this study is represented by the high percentage of missing data (between 50% and 70%) and the high temporal and spatial variability that characterizes the water-quality variables. The competing algorithms implemented belonged to both univariate and multivariate imputation methods (inverse distance weighting (IDW), Random Forest Regressor (RFR), Ridge (R), Bayesian Ridge (BR), AdaBoost (AB), Hubber Regressor (HR), Support Vector Regressor (SVR), and K-nearest neighbors Regressor (KNNR)). According to the results, more than 76% of the imputation outcomes are considered satisfactory (NSE > 0.45). The imputation performance shows better results at the monitoring stations located inside the reservoir than the ones positioned along the mainstream. IDW was the most chosen model for data imputation.

The power system has rapidly grown and expanded over the past decades and has been experiencing major changes and challenges. The increase in energy demand and the modern advancements in the smart grid, such as solar and wind energies and electric vehicles, have led to complexity and complications for utilities. A further layer of complexity and difficulty was added by the rapid expansion of behind-the-meter (BTM) photovoltaic (PV) systems with various designs and characteristic features. The rapid increase and the invisible solar power (BTM) have led to fluctuations in power grid stability and reliability and to inefficiency. Accurate forecasting of load generation will help to assure optimal planning, minimize the negative effects of the PV systems, and minimize the operational and maintenance costs. The authors propose a solution that uses combinations of K-means clustering with neural network machine learning models, AMI real-world PV load generation, and weather data to forecast the generation load at customer locations to achieve a 2.49% error between actual and predicted generation load.

This study is to explore the feasibility using neural network (NN) and deep learning to diagnose three common respiratory diseases with only few symptom words. These three diseases are nasopharyngitis, upper respiratory infection, and bronchitis/bronchiolitis. Through natural language processing, the symptom word vectors are encoded by GPT-2 and classified by the last linear layer of the NN. The experimental results are promising, showing that this model achieves a high performance in predicting all these three diseases. They reach 90% in accuracy, which suggests the implications of the developed model, highlighting its potential use in assisting patients understanding their conditions via a remote diagnosis. Unlike previous studies that focus on extracting various categories of information from medical records, this study directly extracts sequential features from unstructured text data, reducing the effort required for data pre-process.

The working environment of mining electric locomotives is wet and muddy coal mine 1 roadway. There may be idling or slipping between the wheels and rails of mining electric locomotives 2 due to low friction between the wheel and rail and insufficient utilization of creep rate. Therefore, it 3 is necessary to control the creep rate within a reasonable range. In this paper, the autonomous control 4 algorithm for mining electric locomotives based on improved ε-greedy is theoretically proven to be 5 convergent and effective firstly. Secondly, after analyzing the contact state between the wheel and rail 6 under wet and slippery road conditions, it was concluded that the value of creep rate is an important 7 factor affecting the autonomous driving of mining electric locomotives. Therefore, the autonomous 8 control method for mining electric locomotives based on creep control is proposed in this paper. 9 Finally, the effectiveness of the proposed method was verified through simulation. The problem of 10 wheel slipping and idling caused by insufficient friction of mining electric locomotives in coal mining 11 environments is effectively suppressed. Autonomous operation of vehicles with optimal driving 12 efficiency can be achieved through quantitative control and utilization of the creep rate between 13 wheels and rails

Viral infections cause metabolic dysregulation in the infected organism. The present study used metabolomics techniques and machine learning algorithms to retrospectively analyze the alter-ations of a broad panel of metabolites in the serum and urine of a cohort of 126 patients hospi-talized with COVID-19. Results were compared with those of 50 healthy subjects and 45 COVID-19 negative patients but with bacterial infectious diseases. Metabolites were analyzed by gas chro-matography coupled to quadrupole time-of-flight mass spectrometry. The main metabolites al-tered in the sera of COVID-19 patients were those of pentose glucuronate interconversion, ascorbate and fructose metabolism, nucleotide sugars, and nucleotide and amino acid metabolism. Alterations in serum maltose, mannonic acid, xylitol, or glyceric acid metabolites segregated positive patients from the control group with high diagnostic accuracy, while succinic acid seg-regated positive patients from those with other disparate infectious diseases. Increased lauric acid concentrations were associated with severity of infection and death. Urine analyses could not discriminate between groups. Targeted metabolomics and machine learning algorithms facilitated the exploration of the metabolic alterations underlying COVID-19 infection, and to identify po-tential biomarkers for the diagnosis and prognosis of the disease.

Background: In today’s competitive market, predicting customer churn with high accuracy is crucial for enterprises to maintain growth and profitability. Traditional predictive models often lack in accuracy due to the complexity of customer behavior.Objective: This research aims to improve the accuracy of predicting customer churn by utilizing the Particle Swarm Optimization (PSO) algorithm for optimizing the hyperparameters of a composite deep learning model. The performance of this enhanced model is evaluated against traditional models such as LSRM_GRU, LSTM, GRU, and CNN_LSTM to demonstrate the effectiveness of PSO in hyperparameter tuning.Methods: A composite deep learning approach was employed, integrating various neural network architectures to leverage their strengths in modeling complex customer interactions. The PSO algorithm was used to optimize the model’s hyperparameters. Customer transaction and interaction data from different business operations served as the dataset for testing and analyzing the model’s performance. Evaluation metrics including accuracy, precision, recall, F1 score, and ROC AUC were utilized for a detailed comparison with established models.Findings: The PSO-enhanced composite deep learning model showed superior performance across all metrics, significantly outperforming the LSRM_GRU, LSTM, GRU, and CNN_LSTM models. Notably, improvements in ROC AUC and F1 score highlight the robustness and balanced precision-recall trade-off of the proposed model, demonstrating its effectiveness in identifying potential churners.Conclusion: The integration of PSO for hyperparameter optimization in composite deep learning models for customer churn prediction has proven to significantly enhance predictive accuracy and performance metrics over conventional models. This underscores the potential of evolutionary algorithms in improving deep learning applications. Future research should explore the scalability of this approach across various sectors and further refine predictive accuracy with evolving customer data.

The research proposes a self-directed paradigm with self-design using the existing technology with LMS modules, discussions, and self-tests. The research establishes its criteria for assurance of teaching and learning known as 7x2C for the compliance of content, and appearance along with other criteria for the analysis. The above technology of web and LMS is tested for a business course that is technology-oriented in creating an experiential learning system for Covid-19 awareness. Similarly, among other ideas, an environment for educating learners about diabetes and obesity has been designed. The project is known as Sunchoke with its theme of Grow, Eat, and Heal. The educators can use their technology and apply self-directed pedagogy to transform this approach to their customized ways by using their content and rubrics.

This article discusses the comparison between digital and traditional face-to-face coaching within the scope of shadow education institutions. While analyzing the differences and similarities between the two educational models, both their advantages and disadvantages are thoroughly discussed. In this context, interviews were conducted with students and teachers who receive education in both face-to-face and digital coaching, and the positive and negative aspects of both institutions, suitable and unsuitable courses, the future situation, and the effects on students' academic achievements were revealed. According to the results obtained from the research, it is noteworthy that students who do not receive education in digital coaching have prejudices against digitalization. Additionally, no significant difference was found between the academic achievements of students receiving education in digital coaching and those receiving education in face-to-face coaching.

Parkinson’s Disease (PD) is a neurodegenerative disorder characterized by motor and non-motor impairment with various implications on patients’ quality of life. Since currently available therapies are only symptomatic, identifying individuals with prodromal, preclinical or early-stage PD is crucial, as they would be ideal candidates for future disease-modifying therapies. Our analysis aims to develop a robust model for accurate PD detection using accelerometer data collected from PD and non-PD individuals with mild or no tremor during phone conversations. An open-access dataset comprising accelerometer recordings from 22 PD patients and 11 healthy controls (HC) was utilized. The data were pre-processed to extract relevant time, frequency and energy-related features, and a Bidirectional Long Short-Term Memory (Bi-LSTM) model with attention mechanism was employed for classification. The performance of the model was evaluated using 5-fold cross-validation, and metrics of accuracy, precision, recall, specificity, and f1-score were computed. The proposed model demonstrated high accuracy (98%), precision (99%), recall (98%), specificity (96%), and f1-score (98%) in accurately distinguishing PD patients from HC. Our findings indicate that the proposed model outperforms existing approaches and holds promise for detection of PD with subtle symptoms, like tremor, in the wild. Such symptoms can present in the early or even prodromal stage of the disease, and appropriate mobile-based applications may be a practical tool in real-life settings to alert individuals at risk to seek medical assistance or give patients feedback in monitoring their symptoms.

Lung cancer is the leading cause of cancer death worldwide, with non-small cell lung cancer (NSCLC) making up 80% of cases. Some genetic factors leading to NSCLC development include genetic mutations and PD-L1 expression. PD-L1 proteins are targeted in an NSCLC treatment called targeted gene therapy. However, this treatment is effective in a low percentage of patients. This study aimed to create machine learning models to use features like the number of mutations and the level of PD-L1 proteins in cancer cells, along with others, to predict whether a patient will receive clinical benefit from gene therapy treatment. This was done by downloading and merging datasets from cbioportal.org to create a sample size for the model. Features with high correlations to clinical benefit were identified. Three machine-learning models were created using these features to predict clinical benefits in patients, and each model’s accuracy was evaluated. All three models were accurate between 55-85%, with two of the models averaging an accuracy around 75%. Doctors can use these models to more accurately predict whether gene therapy treatment is likely to work in a patient before prescribing it to them.

Currently, skin cancer is the most commonly diagnosed form of cancer in humans and is one of the leading causes of death in patients with cancer. Biopsy methods are an invasive research method and are not always available for primary diagnosis. Imaging methods have low accuracy and depend on the experience of the dermatologist. Artificial intelligence technologies can match and surpass visual analysis methods in accuracy, but they have the risk of a false negative response when a malignant pigmented lesion can be recognized as benign. One possible way to improve accuracy and reduce the risk of false negatives is to analyze heterogeneous data, combine different preprocessing methods, and use modified loss functions to eliminate the negative impact of unbalanced dermatological data. The paper proposes a multimodal neural network system with a modified cross-entropy loss function that is sensitive to unbalanced heterogeneous dermatological data. The accuracy of recognition in 10 diagnostically significant categories for the proposed system was 85.19%. The novelty of the proposed system lies in the use of cross-entropy loss when training the modified function with the help of weight coefficients. The introduction of weighting factors has reduced the number of false negative forecasts, as well as improved accuracy by 1.02-4.03 percentage points compared to the original multimodal systems. The introduction of the proposed multimodal system as an auxiliary diagnostic tool can reduce the consumption of financial and labor resources involved in the medical industry, as well as increase the chance of early detection of skin cancer.

In recent years, advancements in rooftop solar panel technology have sparked a revolution in the electricity markets. This has given rise to a new concept of energy exchange - the ability for consumers and producers to trade localized energy. This concept has been made possible by the emergence of blockchain technology, which has gained significant traction in the energy markets. Its unique ability to facilitate peer-to-peer (P2P) energy transactions has made it a promising solution for the trilemma of scalability, security, and decentralization. However, while blockchain technology has shown great potential, it is still in its early stages of development and has yet to reach its full potential. To fully understand the potential of blockchain technology in P2P energy trading, it is important to explore the trilemma in depth. This study proposes a blockchain-based scalability solution, with a focus on P2P energy trade. This strategy is supported by empirical modeling, utilizing data gathered from a trial case study. The results of this study demonstrate that the suggested technique outperforms base layer models in terms of scalability, while still maintaining the essential elements of security and decentralization. This proposed solution not only has the potential to revolutionize the energy markets, but also has broader implications. By providing a more secure and decentralized platform for P2P energy trading, it has the potential to address issues of energy distribution inequality and promote the adoption of renewable energy. With blockchain technology, individuals and communities have the opportunity to take control of their energy usage and production, reducing their reliance on traditional centralized energy systems. This not only has the potential to lower energy costs for consumers, but also contributes to the overall goal of reducing carbon emissions and mitigating the effects of climate change. The combination of rooftop solar panel technology and blockchain applications has the potential to create a significant shift towards decentralized and sustainable energy systems. This will not only benefit individuals and communities, but also have a positive impact on the environment and the global energy market as a whole. However, for this transition to occur, it is crucial that governments and energy companies continue to support and invest in these advancements.

The accurate prediction of heat load profiles with a daily resolution is required for a broad range of applications such as potential studies, design, or predictive operation of heating systems. If the heat demand of a consumer is mainly caused by (production) processes independent of the ambient temperature, existing load profile prediction methods fail. To close this gap, this study develops two ex post machine learning models for the prediction of the heat demand with a daily resolution. The selected input features are commonly available to each consumer connected to public natural gas and electricity grids or operating an energy monitoring system: Ambient temperature, weekday, electricity consumption, and heat consumption of the last seven days directly before the predicted day. The study’s database covers electricity and natural gas consumption load profiles from 82 German consumers over a period of two years. Electricity and heat consumption correlate strongly with individual patterns for many consumers. Both, shallow and deep learning algorithms are evaluated. The highest accuracy is achieved by a Long Short Term Memory (LSTM) model with a median of R² of 0.94. The ex post model architecture makes the model suitable for automated anomaly detection in energy monitoring systems.

Speech Emotion Recognition (SER) is an interesting and difficult problem to handle. In this paper, we deal with it through the implementation of deep learning networks. We have designed and implemented six different deep learning networks, a Deep Belief Network (DBN), a simple deep neural network (SDNN), a LSTM network (LSTM), a LSTM network with the addition of an attention mechanism (LSTM-ATN), a Convolutional neural network (CNN), and a Convolutional neural network with the addition of an attention mechanism (CNN-ATN), having in mind, apart from solving the SER problem, to test the impact of attention mechanism to the results. Dropout and Batch Normalization techniques are also used to improve the generalization ability (prevention of overfitting) of the models as well as to speed up the training process. The Surrey Audio-Visual Expressed Emotion database (SAVEE), and the Ryerson Audio-Visual Database (RAVDESS) database were used for training and evaluation of our models. The results showed that networks with the addition of the attention mechanism did better than the others. Furthermore, they showed that CNN-ATN was the best among tested networks, achieving an accuracy of 74% for the SAVEE and 77% for the RAVDESS dataset, and exceeded existing state-of-the-art systems for the same datasets.

Despite the importance of few-shot learning, the lack of labeled training data in the real world, makes it extremely challenging for existing machine learning methods as this limited data set does not represent the data variance well. In this research, we suggest employing a generative approach using variational autoencoders (VAEs), which can be used specifically to optimize few-shot learning tasks by generating new samples with more intra-class variations. The purpose of our research is to increase the size of the training data set using various methods to improve the accuracy and robustness of the few-shot face recognition. Specifically, we employ the VAE generator to increase the size of the training data set, including the basic and the novel sets while utilizing transfer learning as the backend. Based on extensive experimental research, we analyze various data augmentation methods to observe how each method affects the accuracy of face recognition. We conclude that the face generation method we proposed can effectively improve the recognition accuracy rate to 96.47% using both the base and the novel sets.

Globalization, knowledge economy, information technology advancements, and the rise of learning societies contribute to the concept of a learning city. The study examines the economic indicators of Yazd city, Iran, in the context of learning city development. A descriptive-analytical approach with a survey method and cluster sampling was employed. Data analysis involved Kolmogorov-Smirnov normality test, single-sample T-test, simple linear regression, and Kendall's W test. The normality test confirmed data normality. The T-test indicated that experts perceived the current state of most variables in Yazd as undesirable. Regression analysis revealed a significant positive association between the economic index and independent variables. Kendall's W test prioritized areas based on economic indicators, with the top 7 and bottom 5 areas identified for focused interventions. The study emphasizes the need to prioritize areas within Yazd based on their economic indicators. This prioritization informs strategies for enhancing social and legal aspects of Yazd's transformation into a learning city. Further research should explore effective methods to cultivate a full-fledged learning environment in the city.

Multimodal remote sensing data classification can enhance the model’s ability to distinguish land features through multimodal data fusion. In this context, how to help models understand the relationship between multimodal data and target tasks is the focus of researchers. Inspired by human feedback learning mechanisms, causal reasoning mechanisms, and knowledge induction mechanisms, this paper integrates causal learning, reinforcement learning, and meta learning into a unified remote sensing data classification framework and proposed the causal meta-reinforcement learning (CMRL). First, based on feedback learning mechanisms, we have overcome the limitations of traditional implicit optimization of fusion features and customized a reinforcement learning environment for multimodal remote sensing data classification tasks. Through feedback interactive learning between agents and the environment, we help them understand the complex relationships between multimodal data and labels, thereby achieving full mining of multimodal complementary information. Second, based on the causal inference mechanism we designed causal distribution prediction action, classification reward, and causal intervention reward, capturing pure causal factors in multimodal data and cutting off false statistical associations between non-causal factors and class labels. Finally, based on the knowledge induction mechanism, we designed a bi-layer optimization mechanism based on meta-learning. By constructing a meta training task and meta validation task simulation model in the generalization scenario of unseen data, we helped the model induce cross-task shared knowledge, thereby improving its generalization ability for unseen multimodal data. The experimental results on multiple sets of multimodal datasets show that the proposed method achieved state-of-the-art performance in multimodal remote sensing data classification tasks.

Online learning is gaining massive popularity with time. The e-learning platforms operate differently from traditional educational institutions and hence need different strategy for course recommendations. This survey aims to cover the major emerging research areas in e-learning recommender systems. Our study covers different areas of research including graph-based methodologies, ITS, query optimization, content-based, and collaborative filtering, big data, and association rules mining. This survey aimed to explore all major emerging directions of recommender systems in education. This study analyzed existing literature in all of the areas mentioned before and performed objective-based analysis. A brief performance analysis was also done for the researches where the values were comparable. Limitations of existing researches were also identified and studied to shed some light on future directions.

Graph neural networks (GNNs) build on the success of deep learning models by extending them for use in graph spaces. Transfer learning has proven extremely successful for traditional deep learning problems: resulting in faster training and improved performance. Despite the increasing interest in GNNs and their use cases, there is little research on their transferability. This research demonstrates that transfer learning is effective with GNNs, and describes how source tasks and the choice of GNN impact the ability to learn generalisable knowledge. We perform experiments using real-world and synthetic data within the contexts of node classification and graph classification. To this end, we also provide a general methodology for transfer learning experimentation and present a novel algorithm for generating synthetic graph classification tasks. We compare the performance of GCN, GraphSAGE and GIN across both the synthetic and real-world datasets. Our results demonstrate empirically that GNNs with inductive operations yield statistically significantly improved transfer. Further we show that similarity in community structure between source and target tasks support statistically significant improvements in transfer over and above the use of only the node attributes.

In healthy humans, taste sensitivity varies widely, influencing food selection and nutritional status. Chemosensory reductions have been associated with numerous pathological disorders or pharmacological interventions. Reliable psychophysical methods are crucial resources to analyze the taste function during routine clinical assessment. However, in the daily clinical routine, they are often considered to be too time-consuming. We used the Supervised Learning (SL) regression method to analyze with high precision the overall taste status of healthy controls (HC) and patients with chemosensory loss and to characterize the combination of responses that best can predict the overall taste status of two groups. Random Forest regressor allowed us to achieve our objective. The analysis of the order of importance and impact of each parameter on the prediction of overall taste status in the two groups showed that salty (low concentration) and sour (high concentration) stimuli specifically characterized healthy subjects, while bitter (high concentration) and astringent (high concentration) stimuli identified patients with chemosensory loss. The identification of these distinctions appears to be of interest to the health system since they may allow the use of specific stimuli during routine clinical assessments of taste function reducing the commitment in terms of time and costs.

Deep learning (DL), a branch of machine learning (ML) and artificial intelligence (AI) is nowadays considered as a core technology of today's Fourth Industrial Revolution (4IR or Industry 4.0). Due to its learning capabilities from data, DL technology originated from artificial neural network (ANN), has become a hot topic in the context of computing, and is widely applied in various application areas like healthcare, visual recognition, cybersecurity, and many more. However, building an appropriate DL model is a challenging task, due to the dynamic nature and variations in real-world problems and data. Moreover, the lack of core understanding turns DL methods into black-box machines that hamper development at the standard level. This article presents a structured and comprehensive view on DL techniques including a taxonomy considering various types of real-world tasks like supervised or unsupervised. In our taxonomy, we take into account deep networks for supervised or discriminative learning, unsupervised or generative learning as well as hybrid learning and relevant others. We also summarize real-world application areas where deep learning techniques can be used. Finally, we point out ten potential aspects for future generation DL modeling with research directions. Overall, this article aims to draw a big picture on DL modeling that can be used as a reference guide for both academia and industry professionals.

During the COVID-19 outbreak, most university courses have been offered on online platforms. A sudden shift from face-to-face classroom learning to online formats could influence the learning effectiveness of classes. This study aims to investigate differences in the learning effectiveness of online and face-to-face lecture courses. It also explores factors that impact the effectiveness of online instruction. These factors include interactions among learners, interactions between learners and the instructor, the quality of online platforms, learners’ ability to use devices and follow instructions, and learners’ situational challenges. The study participants were 261 university students at King Mongkut’s University of Technology Thonburi in Bangkok, Thailand. All participants were enrolled in at least one lecture course, such as history, humans and the environment, the environment and development, or general philosophy, during the 2019 academic year. A questionnaire was distributed to participants after they completed these courses in May 2020. Paired simple t-test analyses were used to compare the effectiveness of online and face-to-face classes, and a multiple regression analysis was used to identify factors that impact the learning effectiveness of online classes. The results show that online classes are less effective than face-to-face courses. The multiple regression analysis also revealed that the effectiveness of online learning was significantly impacted by learners’ ability to interact with classmates during class, their ability to interact with instructors after the class, the quality of online platforms, and disturbances or distractions in learners’ environments.

In recent years, there has been a growing interest in remote sensing image-text cross-modal retrieval due to the rapid development of space information technology and the significant increase in remote sensing image data volume. One approach that has shown promising results in cross-modal retrieval of natural images is the multimodal fusion encoding method. However, remote sensing images have unique characteristics that make the retrieval task challenging. Firstly, the semantic features of remote sensing images are fine-grained, meaning they can be divided into multiple basic units of semantic expression. Additionally, these images exhibit variations in resolution, color, and perspective. Different combinations of basic units of semantic expression can generate diverse text descriptions. These characteristics pose considerable challenges for cross-modal retrieval. To address these challenges, this paper proposes a multi-task guided fusion encoder (MTGFE) based on the multimodal fusion encoding method. The model incorporates three tasks: image-text matching (ITM), masked language modeling (MLM), and the newly introduced multi-view joint representations contrast (MVJRC) task. By jointly training the model with these tasks, we aim to enhance its capability to capture fine-grained correlations between remote sensing images and texts. Specifically, the MVJRC task is designed to improve the model’s consistency in feature expression and fine-grained correlation, particularly for remote sensing images with significant differences in resolution, color, and angle. Furthermore, to address the computational complexity associated with large-scale fusion models and improve retrieval efficiency, this paper proposes a retrieval filtering method. This method achieves higher retrieval efficiency while minimizing accuracy loss. Extensive experiments were conducted on four public datasets to evaluate the proposed method, and the results validate its effectiveness. Overall, this study focuses on remote sensing image-text cross-modal retrieval and introduces the MTGFE model, which combines multimodal fusion encoding with multiple tasks to enhance the model’s ability to capture fine-grained correlations. Additionally, a retrieval filtering method is proposed to improve retrieval efficiency. Experimental results demonstrate the effectiveness of the proposed method.

To assist Java programming learning of novice students, we have developed the web-based Java programming learning assistant system (JPLAS). JPLAS provides several types of exercise problems to cultivate code reading/writing skills at various levels. In JPLAS, the code writing problem (CWP) asks a student to write a source code that will pass the test code given in the assignment where the correctness is verified by running them on JUnit. In this paper, to reduce the teacher load at marking process, we present the answer code validation program that verifies all the source codes from a lot of students for each assignment at a time and reports the number of passing tests for each source code in the CSV file. Besides, to test a source code with various input data, we implement the test data generation algorithm that identifies the data type, generates a new data, and replaces it for each test data in the test code. Furthermore, to verify the correctness of the implemented procedure in the source code, we introduce the intermediate state testing in the test code. For evaluations, we applied the proposal to source codes and test codes in a Java programming course in Okayama university, Japan, and confirmed the validity and effectiveness.

In this paper we propose a computational framework that merges mechanistic modeling with deep neural networks obeying the Systems Biology Markup Language (SBML) standard. Over the last 20 years, the systems biology community has developed a large number of mechanistic models in SBML that are currently stored in public databases. With the proposed framework, existing SBML mechanistic models may be upgraded to hybrid systems through the incorporation of deep neural networks into the model core, using a freely available python tool. The so-formed hybrid mechanistic/neural network models are trained with a deep learning algorithm based on the adaptive moment estimation method (ADAM), stochastic regularization and semidirect sensitivity equations. The trained hybrid models are encoded in SBML and stored back in model databases, where they can be further analyzed as regular SBML models. The application of this approach is illustrated with three well-known case studies: the threonine synthesis model in Escherichia coli, the P58IPK signal transduction model, and the Yeast glycolytic oscillations model. The proposed framework is expected to greatly facilitate the widespread use of hybrid modeling techniques for systems biology applications.

Gene editing tools such as CRISPR-Cas9 have created unprecedented opportunities for genetic studies in plants and animals. We designed a course-based undergraduate research experience (CURE) to train introductory biology students in the concepts and implementation of gene editing technology as well as develop their soft skills in data management and scientific communication. We present two versions of the course that can be implemented with twice-weekly meetings over a five-week period. In the remote-learning version, students perform homology searches, design guide RNAs and primers, and learn the principles of molecular cloning. This version is appropriate when access to laboratory equipment or in-person instruction is limited, such as closures that have occurred in response to the Covid-19 pandemic. In the in-person version, students design guide RNAs, clone CRISPR-Cas9 constructs, and perform genetic transformation of the model plant Arabidopsis thaliana. The highly parallel nature of the CURE makes it possible to target dozens to hundreds of genes, depending on the number of course sections available. Applying this approach in a sensitized mutant background enables focused reverse genetic screens for genetic suppressors or enhancers. The course can be readily adapted to other organisms or projects that employ gene editing.

Automated lip reading is a research problem that has developed considerably in recent years. Lip reading is evaluated both visually and audibly in some cases. The lip reading model is a field of use for detecting specific words using images from security cameras, but it is not possible to use audio-visual databases in this situation. It is not possible to obtain the sound input of the pronounced word in all cases. We collected a new Turkish dataset with only the image in this study. The new dataset is produced using Youtube videos, which is an uncontrolled environment. For this reason, images have difficult parameters in terms of environmental factors such as light, angle, color, and personal characteristics of the face. Despite the different features on the human face such as mustache, beard, and make-up, the visual speech recognition problem was developed on 10 classes including single words and two-word phrases using Convolutional Neural Networks (CNN) without any intervention on the data. The proposed study using only-visual data obtained a model which is automated visual speech recognition with a deep learning approach. In addition, since this study uses only-visual data, the computational cost and resource usage is less than in multi-modal studies. It is also the first known study to address the lip reading problem with a deep learning algorithm using a new dataset belonging to the Ural-Altaic languages.

Machine learning is an established and frequently used technique in industry and academia but a standard process model to improve success and efficiency of machine learning applications is still missing. Project organizations and machine learning practitioners have a need for guidance throughout the life cycle of a machine learning application to meet business expectations. We therefore propose a process model for the development of machine learning applications, that covers six phases from defining the scope to maintaining the deployed machine learning application. The first phase combines business and data understanding as data availability oftentimes affects the feasibility of the project. The sixth phase covers state-of-the-art approaches for monitoring and maintenance of a machine learning applications, as the risk of model degradation in a changing environment is eminent. With each task of the process, we propose quality assurance methodology that is suitable to address challenges in machine learning development that we identify in form of risks. The methodology is drawn from practical experience and scientific literature and has proven to be general and stable. The process model expands on CRISP-DM, a data mining process model that enjoys strong industry support but lacks to address machine learning specific tasks. Our work proposes an industry and application neutral process model tailored for machine learning applications with focus on technical tasks for quality assurance.

This work presents significant results obtained by the application of global optimization techniques to the design of finite, normal form games with mixed strategies. To that end, the Fuzzy ASA global optimization method is applied to several design examples of strategic games, demonstrating its effectiveness in obtaining payoff functions whose corresponding games present a previously established Nash equilibrium. In other words, the game designer becomes able to choose a convenient Nash equilibrium for a generic finite state strategic game and the proposed method computes payoff functions that will realize the desired equilibrium, making it possible for the players to reach the favorable conditions represented by the chosen equilibrium. Considering that game theory is a very significant approach for modeling interactions between competing agents, and Nash equilibrium represents a powerful solution concept, portraying situations in which joint strategies are optimal in the sense that players cannot benefit from individually modifying their current strategies provided that other players do not change their strategies as well, it is natural to infer that the proposed method may be very useful for strategists in general. In summary, it is a genuine instance of artificial inference of payoff functions after a process of global machine learning, applied to their numerical components.

Species distribution modeling (SDM) has come a long way since its inception. Starting as simple bioclimatic envelope models based on expert knowledge, species distribution models (SDMs) have evolved into complex and sophisticated models that incorporate multiple sources of data and machine learning algorithms. Today, SDMs play a crucial role in addressing pressing conservation and management issues, including the impacts of climate change on species ranges and the as-sessment of species vulnerability to extinction. In this article, we will embark on a journey through the history, present, and future of SDM, exploring its evolution from bioclimatic envelopes to machine learning. We will also provide practical tips on how to use SDMs effectively and discuss the exciting future developments in this field. Whether you are a seasoned SDM expert or new to this field, this article will provide valuable insights into the exciting world of SDM. By exploring the rich history and current state of the field, we hope to shed light on the tremendous potential of SDM for improving our understanding of the distribution of species in a changing world.

Electroencephalography (EEG) signals are the primary source for discriminating the preictal from the interictal stage, enabling early warnings before the seizure onset. The epileptic seizure prediction models face significant challenges due to data scarcity, diversity, and privacy. This paper proposes a three-tier architecture for epileptic seizure prediction associated with the Federated Learning (FL) model that enhances the capability by utilizing the significant amount of seizure patterns from globally distributed patients with data privacy. The determination of the preictal state is influenced by global and local model-assisted decision-making by modeling the two-level edge layer. Integrating the Spiking Encoder (SE) with Graph Convolutional Neural Network (Spiking-GCNN) works as the local model trained using the bi-timescale approach. Each local model utilizes the aggregated seizure knowledge from the different medical centers through FL and determines the preictal probability in the coarse-grained personalization. The Adaptive Neuro-Fuzzy Inference System (ANFIS) is utilized in fine-grained personalization to recognize epileptic seizure patients by examining the outcomes of the FL model, heart rate variability features, and patient-specific clinical features. The proposed seizure prediction is evaluated using benchmark datasets by comparing them with the existing works to demonstrate the potential results.

This paper describes and demonstrates an autonomous robotic team that can rapidly learn the characteristics of environments that it has never seen before. The flexible paradigm is easily scalable to multi-robot, multi-sensor autonomous teams, and is relevant to satellite calibration/validation and the creation of new remote sensing data products. A case study is described for the rapid characterisation of the aquatic environment, over a period of just a few minutes we acquired thousands of training data points. This training data allowed our machine learning algorithms to rapidly learn by example and provide wide area maps of the composition of the environment. Along side these larger autonomous robots two smaller robots that can be deployed by a single individual were also deployed (a walking robot and a robotic hover-board), observing significant small scale spatial variability.

Aim: To determine whether an AI model and single sensor measuring acceleration and ECG could model cognitive and physical fatigue for a self-paced trail run. Methods: A field-based protocol of continuous fatigue repeated hourly induced physical (~45 minutes) and cognitive (~10 minutes) fatigue on one healthy participant. Physical load was a 3.8 km, 200 m vertical gain, trail run with acceleration and electrocardiogram (ECG) data collected using a single sensor. Cognitive load was a Multi Attribute Test Battery (MATB) and separate assessment battery including the Finger Tap Test (FTT), Stroop, Trail Making A and B, Spatial Memory, Paced Visual Serial Addition Test (PVSAT), and a vertical jump. A fatigue prediction model was implemented using a Convolutional Neural Network (CNN). Results: When the fatigue test battery results were compared for sensitivity to the protocol load, FTT right hand (R2 0.71) and Jump Height (R2 0.78) were the most sensitive while the other tests were less sensitive (R2 values Stroop 0.49, Trail Making A 0.29, Trail Making B 0.05, PVSAT 0.03, spatial memory 0.003). Best prediction results were achieved with a rolling average of 200 predictions (102.4 s), during set activity types, mean absolute error for ‘walk up’ (MAE200 12.5%) and range of absolute error for ‘run down‘ (RAE200 16.7%). Conclusion: We were able to measure cognitive and physical fatigue using a single wearable sensor during a practical field protocol including contextual factors in conjunction with a neural network model. This research has practical application to fatigue research in the field.

Interest in drone solutions in forestry applications is growing. Using drones, datasets can be captured flexibly and at high spatial and temporal resolutions when needed. In forestry applications, fundamental tasks include the detection of individual trees, tree species classification, bio-mass estimation, etc. Deep Neural Networks (DNN) have shown superior results when comparing with conventional machine learning methods such as Multi-Layer Perceptron (MLP) in cases of huge input data. The objective of this research was to investigate 3D convolutional neural networks (3D-CNN) to classify three major tree species in a boreal forest: pine, spruce, and birch. The proposed 3D-CNN models were employed to classify tree species in a test site in Finland. The classifiers were trained with a dataset of 3039 manually labelled trees. Then the accuracies were assessed by employing independent datasets of 803 records. To find the most efficient set of feature combination, we compare the performances of 3D-CNN models trained with hyperspectral (HS) channels, RGB channels, and canopy height model (CHM), separately and combined. It is demonstrated that the proposed 3D-CNN model with RGB and HS layers produces the highest classification accuracy. The producer accuracy of the best 3D-CNN classifier on the test dataset were 99.6%, 94.8%, and 97.4% for pines, spruces, and birches, respectively. The best 3D-CNN classifier produced ~5% better classification accuracy than the MLP with all layers. Our results suggest that the proposed method provides excellent classification results with acceptable performance metrics for HS datasets. Our results show that pine class was detectable in most layers. Spruce was most detectable in RGB data, while birch was most detectable in the HS layers. Furthermore, the RGB datasets provide acceptable results for many low-accuracy applications.

The accurate forecasting of solar power output is essential for ensuring the stable and efficient operation of grid-connected photovoltaic (PV) systems, especially when faced with variable weather patterns. This research focuses on the integration of Machine Learning (ML) methodologies and climatic parameters to predict solar panel energy generation, with a specific emphasis on addressing consumption-production imbalances. Leveraging a dataset sourced from the Kaggle platform, the study is conducted in the context of Estonia, aiming to optimize solar energy utilization in this geographic region. The dataset, obtained from Kaggle, encompasses comprehensive information on climatic variables, including sunlight intensity, temperature, and humidity, alongside corresponding solar panel energy output. Leveraging machine learning algorithms, including XGBoost regression and neural networks, our model aims to identify complex patterns and relationships within the datasets. By tailoring the model to Estonia's climatic nuances, we seek to enhance the accuracy of energy production forecasts and, consequently, better manage the challenges associated with consumption-production imbalances. Furthermore, the research investigates the adaptability of the proposed model to diverse climatic conditions, ensuring its applicability for similar endeavors in other geographical locations. The performance of the forecasting model will be evaluated using error metrics like mean absolute error (MAE) and root mean squared error (RMSE). This study leverages a comprehensive dataset from Kaggle and utilizes advanced machine learning techniques to generate valuable insights. These insights can inform the development of sustainable energy policies and practices, ultimately leading to a more efficient and reliable renewable energy infrastructure.

Learning disorders (LD) are diagnosed in children impaired in the academic skills of reading, writing and/or mathematics. Children with LD usually show a slower resting-state electroencephalogram (EEG), with EEG patterns corresponding to a neurodevelopmental lag. LD-children also show a consistent cognitive impairment in working memory (WM), including an abnormal task-related EEG with an overall slower EEG activity of more delta and theta power, and less gamma activity in posterior sites; task-related EEG patterns considered indices of an inefficient neural resource management. Neurofeedback (NFB) treatments aimed at normalizing the resting-state EEG of LD-children have shown improvements in cognitive-behavioral indices and diminished EEG abnormalities. Given the typical findings of a WM impairment in LD-children; we aimed to explore the effects of a NFB treatment in the WM of children with LD, by analyzing the WM-related EEG power-spectrum. We recruited 18 children with LD (8-10 years old). They performed a Sternberg-type WM-task synchronized with an EEG of 19 leads (10-20 system) twice in pre-post treatment conditions. They went through either 30 sessions of a NFB treatment (NFB-group, n= 10); or through 30 sessions of a placebo-sham treatment (Sham-group, n= 8). We analyzed the before-after treatment group differences for the behavioral performance and the WM-related power-spectrum. The NFB group showed faster response times in the WM-task post-treatment. They also showed an increased gamma power at posterior sites and a decreased beta power. We explain these findings in terms of NFB improving the maintenance of memory representations coupled with a reduction of anxiety.

Extra virgin olive oil (EVOO) is the highest quality of olive oil and is characterized by highly beneficial nutritional properties. The large increase in both consumption and fraud, for example through adulteration, creates new challenges and an increasing demand for developing new quality assessment methodologies that are easier and cheaper to perform. As of today, the determination of olive oil quality is performed by producers through chemical analysis and organoleptic evaluation. The chemical analysis requires the advanced equipment and chemical knowledge of certified laboratories, and has therefore a limited accessibility. In this work a minimalist, portable and low-cost sensor is presented, which can perform olive oil quality assessment using fluorescence spectroscopy. The potential of the proposed technology is explored by analyzing several olive oils of different quality levels, EVOO, virgin olive oil (VOO), and lampante olive oil (LOO). The spectral data were analyzed using a large number of machine learning methods, including artificial neural networks. The analysis performed in this work demonstrates the possibility of performing classification of olive oil in the three mentioned classes with an accuracy of 100%. These results confirm that this minimalist low-cost sensor has the potential of substituting expensive and complex chemical analysis.

Fruit quality is a critical factor in the produce industry, affecting producers, distributors, consumers, and the economy. High-quality fruits are more appealing, nutritious, and safe, boosting consumer satisfaction and revenue for producers. Artificial Intelligence can aid in assessing the quality of the fruit using images. This paper presents a general machine-learning model for assessing fruit quality using deep image features. The model leverages the learning capabilities of the recent successful networks for image classification called Vision Transformers (ViT). The ViT model is built and trained with a combination of various fruit datasets and learned to distinguish between good and rotten fruit images. The general model demonstrated impressive results in accurately identifying the quality of various fruits such as Apples (with a 99.50% accuracy), Cucumbers (99%), Grapes (100%), Kakis (99.50%), Oranges (99.50%), Papayas (98%), Peaches (98%), Tomatoes (99.50%), and Watermelons (98%). However, it showed slightly lower performance in identifying Guavas (97%), Lemons (97%), Limes (97.50%), mangoes (97.50%), Pears (97%), and Pomegranates (97%).

Over the past decade, research in the field of Deep Learning has brought about novel improvements in image generation and feature learning; one such example being a Generative Adversarial Network. However, these improvements have been coupled with an increasing demand on mathematical literacy and previous knowledge in the field. Therefore, in this literature review, I seek to introduce Generative Adversarial Networks (GANs) to a broader audience by explaining their background and intuition at a more foundational level. I begin by discussing the mathematical background of this architecture, specifically topics in linear algebra and probability theory. I then proceed to introduce GANs in a more theoretical framework, along with some of the literature on GANs, including their architectural improvements and image-generation capabilities. Finally, I cover state-of-the-art image generation through style-based methods, as well as their implications on society.

This study evaluated the using of machine vision in combination with deep learning to identify weeds in real-time for wheat production system. PMAS-Arid Agriculture university research farm were selected for collection of images (6000 total images) of weeds and wheat crops under different weather condition. During growing season, the database was constructed to identify the weeds. For this study two framework were used TensorFlow and PyTorch under CNNs and Deep learning. Deep learning perfromed better with in PyTourch value as compared to another model in Tensorflow. comparing with other networks such as YOLOv4, we concluded that our network reaches a better result between speed and accuracy. Specifically, the maximum precision of weed and wheat plants were 0.89 and 0.91 respectively with 9.43 ms and 12.38 ms inference time per image (640 × 640) NVIDIA RTX2070 GPU.

Tremendous progress has been witnessed in artificial intelligence within the domain of neural network backed deep learning systems and its applications. As we approach the post Moore’s Law era, the limit of semiconductor fabrication technology along with a rapid increase in data generation rates have lead to an impending growing challenge of tackling newer and more modern machine learning problems. In parallel, quantum computing has exhibited rapid development in recent years. Due to the potential of a quantum speedup, quantum based learning applications have become an area of significant interest, in hopes that we can leverage quantum systems to solve classical problems. In this work, we propose a quantum deep learning architecture; we demonstrate our quantum neural network architecture on tasks ranging from binary and multi-class classification to generative modelling. Powered by a modified quantum differentiation function along with a hybrid quantum-classic design, our architecture encodes the data with a reduced number of qubits and generates a quantum circuit, loading it onto a quantum platform where the model learns the optimal states iteratively. We conduct intensive experiments on both the local computing environment and IBM-Q quantum platform. The evaluation results demonstrate that our architecture is able to outperform Tensorflow-Quantum by up to 12.51% and 11.71% for a comparable classic deep neural network on the task of classification trained with the same network settings. Furthermore, our GAN architecture runs the discriminator and the generator purely on quantum hardware and utilizes the swap test on qubits to calculate the values of loss functions. In comparing our quantum GAN, we note our architecture is able to achieve similar performance with 98.5% reduction on the parameter set when compared to classical GANs. With the same number of parameters, additionally, QuGAN outperforms other quantum based GANs in the literature for up to 125.0% in terms of similarity between generated distributions and original data sets.

The conventional machine learning (ML) algorithms are continuously advancing and evolving at a fast-paced by introducing the novel learning algorithms. ML models are continually improving using hybridization and ensemble techniques to empower computation, functionality, robustness, and accuracy aspects of modeling. Currently, numerous hybrid and ensemble ML models have been introduced. However, they have not been surveyed in a comprehensive manner. This paper presents the state of the art of novel ML models and their performance and application domains through a novel taxonomy.

This paper presents a new method : HIVEC to learn latent vector representations of graphs in a manner that captures the semantic dependencies of sub-structures. The representations can then be used in machine learning algorithms for tasks such as graph classification, clustering etcetera. The method proposed is unsupervised and uses the information of co-occurrence of sub-structures. It introduces a notion of hierarchical embeddings that allows us to avoid repetitive learning of sub-structures for every new graph. As an alternative to deep learning methods, the edit distance similarity between sub-structures is also used to learn vector representations. We compare the performance of these methods against previous work.

Undergraduate learning is a multifaceted concept that many researchers have investigated to promote student engagement, academic performance, etc. Exploring factors promoting learning among undergraduates in FYW programs, which are required for undergraduates might potentially contribute to sustainable, cross-disciplinary undergraduate education. This study used the 10-year longitudinal grade and survey data at San José State University (SJSU) and the data was collected from Academic Years (AYs) 2015 to 2020, and five student surveys constituted the sample size of 18,101 students. In 2016, SJSU changed the remedial writing course to college writing courses. Since then, students need to complete an online module course before placing themselves into a one-semester or two-semester writing course. The previous study showed that students with higher writing comfort and confidence tend to place themselves into a semester-long class. Regression analysis was applied to examine the relationships among factors promoting learning among those students. University admitted students from various diversity and backgrounds, the results of this study suggested that what students chose and persisted in were more important in advancing their academic success rather than prior learning experience or psychological factors before the class.

The COVID-19 pandemic has led to the growth of hybrid and online learning environments and the trend to introduce more technology into the classroom. One such change would be the use of smart synchronous hybrid learning environments (SSHLE), which are settings with both in-person and online students concurrently, and in which technology plays a key role in sensing, analyzing, and reacting throughout the teaching and learning process. These changing environments and the incorporation of new technologies can place a greater orchestration load on participants and a reduction in teacher agency. In this context, the aim of this paper is to analyse the orchestration load and teacher agency in different SSHLEs. The NASA-TLX model was used to measure the orchestration load in several scenarios. Questionnaires and interviews were used to measure teacher agency. The results obtained indicate that the orchestration load of the teacher tends to be high (between 60 and 70 points out of 100 of the NASA-TLX workload), especially when they lack experience in synchronous hybrid learning environments, and the orchestration load of the students tends to have average values (between 50 and 60) in the SSHLEs analysed. Meanwhile, the teacher agency does not appear to be altered but shows potential for improvement.

A common challenge in real-world classification scenarios with sequentially appending target domain data is insufficient training datasets during the training phase. Therefore, conventional deep learning and transfer learning classifiers are not applicable especially when individual classes are not represented or are severely underrepresented at the outset. Domain Generalization approaches reach their limits when domain shifts become too large, making them occasionally unsuitable as well. In many (technical) domains, however, it is only the defect/ worn/ reject classes that are insufficiently represented, while the non-defect class is often available from the beginning. The proposed classification approach addresses such conditions and is based on a CNN encoder. Following a contrastive learning approach, it is trained with a modified triplet loss function using two datasets: Besides the non-defective target domain class (= 1st dataset), a state-of-the-art labeled source domain dataset that contains highly related classes (e.g., a related manufacturing error or wear defect) but originates from a (highly) different domain (e.g., different product, material, or appearance) (= 2nd dataset) is utilized. The approach learns the classification features from the source domain dataset while at the same time learning the differences between the source and the target domain in a single training step, aiming to transfer the relevant features to the target domain. The classifier becomes sensitive to the classification features and – by architecture – robust against the highly domain-specific context. The approach is benchmarked in a technical and a non-technical domain and shows convincing classification results. In particular, it is shown that the domain generalization capabilities and classification results are improved by the proposed architecture, allowing for larger domain shifts between source and target domains.

A patient will visit physicians when he/she feels ill. This illness is not for COVID-19 but it is a general tendency of human being to visit doctor probably it can not be controlled by general drug. When a patient comes to a doctor, the doctor examines him/her after knowing his/her problem. The physician always asks him/her about some questions related to him/her daily life. For example, if a young male patient comes to a doctor with a symptom of fever and cough, the first question doctor asked him that he has a habit of smoking. Then doctor asks him whether this type of symptom appeared often to him previously or not. If the answers of both questions are yes, then the first one is habit and the second one is that he may suffering from some serious disease or a disease due to the weather. The aim of this paper is to consider habit of the patient as well as he/she has been affected by a critical disease. This information is used to build a model that will predict whether there is any possibility of his/her being affected by COVID-19. This research work contributes to tackle the pandemic situation occurred due to Corona Virus Infectious Disease, 2019 (Covid-19). Outbreak of this disease happens based on numerous factors such as past health records and habits of patients. Health records include diabetes tendency, cardiovascular disease existence, pregnancy, asthma, hypertension, pneumonia; chronic renal disease may contribute to this disease occurrence. Past lifestyles such as tobacco, alcohol consumption may be analyzed. A deep learning based framework is investigated to verify the relationship between past health records, habits of patients and covid-19 occurrence. A stacked Gated Recurrent Unit (GRU) based model is proposed in this paper that identifies whether a patient can be infected by this disease or not. The proposed predictive system is compared against existing benchmark Machine Learning classifiers such as Support Vector Machine (SVM) and Decision Tree (DT).

Count data analysis presents unique challenges due to its discrete nature, often exhibiting excessive zeros and overdispersion. To address these complexities, count models, such as Poisson regression and Negative Binomial regression, have been developed, enabling modeling and prediction of count-based phenomena. Additionally, it’s important to notice that zero inflation, a phenomenon commonly observed in count data, requires specialized techniques for robust analysis. This article provides an overview of count data and count models, explores zero inflation, introduces Likelihood Ratio Tests, and explains how the Vuong Test can be used as a model selection criteria. Furthermore, we created a Vuong Test implementation from scratch using the Python programming language. This implementation enhances the accessibility and applicability of the Vuong Test in real-world scenarios, providing a valuable contribution to the academic community, since Python didn’t have an implementation of this statistical test.

Proxy re-encryption (PRE) is a momentous and widely used cryptographic technique. It enables a proxy to forward ciphertext without the need of decryption. PRE has received significant interest in applications like cloud computing, blockchain, and the Internet of Things. Despite its wide range of uses, PRE has also been subject to new security and privacy regulations. In PKC’19, Cohen $et~al.$~first drew attention to the weakness in PRE's security against chosen-plaintext attacks (CPA) and put up a more stringent security concept known as security against honest re-encryption attacks (HRA). Notably, Cohen provides a beneficial conclusion as well. It is proveded that PRE schemes with re-encryption simulatability property can be elevated from CPA to HRA security. It is also proved that CPA-secure PRE schemes with re-encryption simulatability property can be directly elevated to those satisfying HRA security. However, those PRE schemes with re-encryption simulatability are almost always based on pairings. In this study, to the best of our knowledge, we directly construct HRA-secure PRE with re-encryption simulatability for the first time based on the learning with errors (LWE) assumption, which was widely believed to be quantum-resistant. Based on the re-encryption key generation algorithm and the re-encryption algorithm construction method of the above scheme, we can modify attribute-based conditional proxy re-encryption (AB-CPRE) as well as the corresponding attribute-based proxy re-encryption (AB-PRE) algorithm to make them have re-encryption simulatability properties. Finally, by using this property, We boost the security of AB-CPRE scheme of ESORICS'21 from CPA to HRA and simplify the HRA-security proof for the AB-PRE of ESORICS'21.

The primary barrier to understanding visual and abstract information in STEM fields is representational competence the ability to generate, transform, analyze and explain representations. The relationship is known between the foundational visual literacy and the domain specific science literacy, however how science literacy is a function of science learning is still not well understood despite investigation across many fields. To support the improvement of students’ representational competence and promote learning in science, identification of visualization skills is necessary. This project details the development of an artificial neural network (ANN) capable of measuring and modeling visual science literacy (VSL) via neurological measurements using functional near infrared spectrometry (fNIRS). The developed model has the capacity to classify levels of scientific visual literacy allowing educators and curriculum designers the ability to create more targeted and immersive classroom resources such as virtual reality, to enhance the fundamental visual tools in science.

We introduce a modelling framework for the investigation of on-line machine learning processes in non-stationary environments. We exemplify the approach in terms of two specific model situations: In the first, we consider the learning of a classification scheme from clustered data by means of prototype-based Learning Vector Quantization (LVQ). In the second, we study the training of layered neural networks with sigmoidal activations for the purpose of regression. In both cases, the target, i.e. the classification or regression scheme, is considered to change continuously while the system is trained from a stream of labeled data. We extend and apply methods borrowed from statistical physics which have been used frequently for the exact description of training dynamics in stationary environments. Extensions of the approach allow for the computation of typical learning curves in the presence of concept drift in a variety of model situations. First results are presented and discussed for stochastic drift processes in classification and regression problems. They indicate that LVQ is capable of tracking a classification scheme under drift to a non-trivial extent. Furthermore, we show that concept drift can cause the persistence of sub-optimal plateau states in gradient based training of layered neural networks for regression.

. Identifying students who might have difficulty in their course of studies ahead of time is crucial. There can be many reasons for performance issues, such as personality, family, social, and/or economic. We advocate that educational systems should use machine learning to predict students’ performance based on performance factors. This would allow educational professionals and institutions to put in place a preventive plan to help students towards achievements of their educational goals and success. In this chapter, we propose a student performance prediction method and evaluate its performance. We provide a taxonomy of performance factors that help to gauge students performance from different perspectives and give insights on the categories and features that have more significant impact on students’ performance. The results of this work can be used by education institutions to put in place a student-centric approach to tackle performance issues before they create long-term effects on student’s life. In addition, it will help education policymakers to introduce a tailored approach for the population in specific areas.

Pattern recognition techniques form the heart of most, if not all, incoherent linear shift-invariant systems. When an object is recorded using a camera, the object information gets sampled by the point spread function (PSF) of the system, replacing every object point with the PSF in the sensor. The PSF is a sharp Kronecker Delta-like function when the numerical aperture (NA) is large with no aberrations. When the NA is small, and the system has aberrations, the PSF appears blurred. In the above case, if the PSF is known, then the object information can be obtained by scanning the PSF over the recorded object intensity pattern and looking for pattern matching conditions through a mathematical process called correlation. In this study, a recently developed deconvolution method, the Lucy-Richardson-Rosen algorithm (LR2A), has been implemented to computationally refocus images recorded in the presence of spatio-spectral aberrations. The performance of LR2A was compared against the Lucy-Richardson algorithm and non-linear reconstruction. LR2A exhibits a superior deconvolution capability even in extreme cases of spatio-spectral aberrations and blur. Experimental results of deblurring a picture captured using high-resolution smartphone cameras are presented. LR2A was implemented to significantly improve the performances of the widely used deep convolutional neural networks for image classification.

Cities are undergoing huge shifts in technology and operations in recent days, and `data science' is driving the change in the current age of the Fourth Industrial Revolution (Industry 4.0 or 4IR). Extracting insights or actionable knowledge from city data and building a corresponding data-driven model is the key to making a city system automated and intelligent. Data science is typically the study and analysis of actual happenings with historical data using a variety of scientific methodology, machine learning techniques, processes, and systems. In this paper, we concentrate on and explore ``Smart City Data Science", where city data collected from various sources like sensors and Internet-connected devices, is being mined for insights and hidden correlations to enhance decision-making processes and deliver better and more intelligent services to citizens. To achieve this goal, various machine learning analytical modeling can be employed to provide deeper knowledge about city data, which makes the computing process more actionable and intelligent in various real-world services of today's cities. Finally, we identify and highlight ten open research issues for future development and research in the context of data-driven smart cities. Overall, we aim to provide an insight into smart city data science conceptualization on a broad scale, which can be used as a reference guide for the researchers, professionals, as well as policy-makers of a country, particularly, from the technological point of view.

Accurate geometric modeling of blood vessels lumen given their 3D images is a basis for vessel quantification in support of cardiovascular diseases diagnosis, treatment and monitoring. Unlike other approaches, which assume circular or elliptical vessel cross-section, the proposed method employs parametric B-splines combined with the image formation system equations to accurately localize the highly curved lumen boundaries. No image segmentation is involved which might reduce the localization accuracy in effect of spatial discretization. It is shown that the model parameters can be reliably identified by a feedforward neural network which, driven by the cross-section images, predicts the parameter values many times faster than a reference least-squares (LS) model fitting algorithm. Two example applications: modeling the lower extremities artery-vein complexes visualized in steady-state contrast-enhanced magnetic resonance images (MRI) and the coronary arteries pictured in computed tomography angiograms (CTA), are presented and discussed to demonstrate the method potential. Besides application to medical diagnosis, blood-flow simulation and vessel-phantom design, it can serve as a tool for automated annotation of image datasets required to train machine-learning algorithms.

Sparse training (ST) aims to ameliorate deep learning by replacing fully connected artificial neural networks (ANNs) with sparse or ultra-sparse ones, such as brain networks are, therefore it might benefit to borrow brain-inspired learning paradigms from complex network intelligence theory. Here, we launch the ultrasparse advantage challenge, whose goal is to offer evidence on the extent to which ultra-sparse (around 1% connection retained) topologies can achieve any leaning advantage on fully connected. Epitopological learning is a field of network science and complex network intelligence that studies how to implement learning on complex networks by changing the shape of their connectivity structure (epitopological plasticity). One way to implement Epitopological (epi- means new) Learning is via link prediction: predicting the likelihood of nonobserved links to appear in the network. Cannistraci-Hebb learning theory inspired the CH3-L3 network automata rule for link prediction which is effective for general-purpose link prediction. Here, starting from CH3-L3 we propose Epitopological Sparse Meta-deep Learning (ESML) to apply Epitopological Learning to sparse training. In empirical experiments, we find that ESML learns ANNs with ultra-sparse hyperbolic (epi-)topology in which emerges a community layer organization that is meta-deep (meaning that each layer also has an internal depth due to powerlaw node hierarchy). Furthermore, we discover that ESML can in many cases automatically sparse the neurons during training (arriving even to 30% neurons left in hidden layers), this process of node dynamic removal is called percolation. Starting from this network science evidence, we design Cannistraci-Hebb training (CHT), a 4-step training methodology that put ESML at its heart. We conduct experiments on 6 datasets and 3 network structures (MLPs, VGG16, ResNet50) comparing CHT to dynamic sparse training SOTA algorithms and fully connected network. The results indicate that, with a mere 1% of links retained during training, CHT surpasses fully connected networks on VGG16 and ResNet50. This key finding is an evidence for ultra-sparse advantage and signs a milestone in deep learning. CHT acts akin to a gradient-free oracle which adopts CH3-L3 based epitopological learning to guide the placement of new links in the ultra-sparse network topology to facilitate sparse-weight gradient learning, and this in turn reduces the convergence time of ultra-sparse training. Finally, CHT offers first examples of parsimony dynamic sparse training because, in many datasets, it canretain network performance by percolating and significantly reducing the node network size.