The COVID-19 pandemic affected the whole world, but not all countries were impacted equally. This opens the question of what factors can explain the initial faster spread in some countries compared to others. Many such factors are overshadowed by the effect of the countermeasures, so we studied the early phases of the infection when countermeasures have not yet taken place. We collected the most diverse dataset of potentially relevant factors and infection metrics to date for this task. Using it, we show the importance of different factors and factor categories as determined by both statistical methods and machine learning (ML) feature selection (FS) approaches. Factors related to culture (e.g., individualism, openness), development, and travel proved the most important. A more thorough factor analysis was then made using a novel rule discovery algorithm. We also show how interconnected these factors are and caution against relying on ML analysis in isolation. Importantly, we explore potential pitfalls found in the methodology of similar work and demonstrate their impact on COVID-19 data analysis. Our best models using the decision tree classifier can predict the infection class with roughly 80% accuracy.

Skin cancer is an exquisite disease globally nowadays. Because of the poor contrast and apparent resemblance between skin and lesions, automatic identification of skin cancer is complicated. The rate of human death can be massively reduced if melanoma skin cancer can be detected quickly using dermoscopy images. In this research, an anisotropic diffusion filtering method is used on dermoscopy images to remove multiplicative speckle noise and the fast-bounding box (FBB) method is applied to segment the skin cancer region. Furthermore, the paper consists of two feature extractor parts. One of the two features extractor parts is the hybrid feature extractor (HFE) part and another is the convolutional neural network VGG19 based CNN feature extractor part. The HFE portion combines three feature extraction approaches into a single fused feature vector: Histogram-Oriented Gradient (HOG), Local Binary Pattern (LBP), and Speed Up Robust Feature (SURF). The CNN method also is used to extract additional features from test and training datasets. This two-feature vector is fused to design the classification model. This classifier performs the classification of dermoscopy images whether it is melanoma or non-melanoma skin cancer. The proposed methodology is performed on two ordinary datasets and achieved the accuracy 99.85%, sensitivity 91.65%, and specificity 95.70%, which makes it more successful than previous machine learning algorithms.

This study compares the performance of existing studies on multimodal emotion recognition, and proposes a model that fuses two modalities with the speaker's text and voice signals as input values and detects depression. Based on the DAIC-WOZ dataset, voice features were extracted using CNN, text features were extracted using Transformers, and two modalities were fused through a tensor fusion network. We also build a model to detect whether the speaker is depressed or not using LSTM in the final layer. This study suggests the possibility of increasing access to mental illness diagnosis by enabling patients to detect depression on their own in daily conversations. If the model proposed in this study is developed and the voice conversation system is connected, it will be easier for patients who cannot visit the hospital periodically or who are reluctant to visit the hospital to check their condition and seek recovery. Furthermore, it can be expanded to multi-label classification for various mental diseases and used as a simple self-mental disease diagnosis tool.

Caries is a prevalent oral disease that primarily affects children and teenagers. Advances in ma-chine learning have caught the attention of scientists working with decision support systems to predict early tooth decay. Current research has developed machine learning algorithm for caries classification and reached high accuracy especially in ML for image data. Unfortunately, most studies on dental caries only focus on classification and prediction tasks, meanwhile dental carries prevention is more important. Therefore, this study aims to design an efficient feature for decision support system machine learning based that can identify various risk factors that cause dental caries and its prevention. The data used in the research work was obtained from the 2018 Korean Children's Oral Health Survey, which totaled nine datasets. The experimental results show that combining the mRMR and GINI Feature Importance methods when training with the GBDT model achieved the optimum performance of 95%, 93%, 99%, and 88% for accuracy, F1 score, precision, and recall, respectively. So, the proposed method has provided effective predictive model for dental caries prediction.

Electrical rotating machines like Induction Motors (IMs) are widely used in several industrial applications since their robust elements, provide high efficiency and give versatility in industrial applications. Nevertheless, the occurrence of faults in IMs is inherent to their operating conditions, hence, Inter-turn short-circuit (ITSC) is one of the most common failures that affect IMs and its appearance is due to electrical stresses leads to the degradation of the stator winding insulation. In this regard, this work proposes a diagnosis methodology for the assessment and detection of incipient ITSC in IMs, the proposed method is based on the processing of vibration, stator currents and magnetic stray-flux signals. Certainly, the novelty and contribution include the characterization of different physical magnitudes by estimating a set of statistical time domain features, as well as, their fusion and reduction through the Linear discriminant Analysis technique within a feature-level fusion approach. Furthermore, the fusion and reduction of information from different physical magnitudes leads to perform the automatic fault detection and identification by a simple Neural-Network (NN) structure. The proposed method is evaluated under a complete set of experimental data and the obtained results demonstrate that the fusion of information from different sources (physical magnitudes) allows to improve the accuracy during the detection of ITSC in IMs , the results make this proposal feasible to be incorporated as a part of condition-based maintenance programs in the industry.

With the evolution of computing technology in many application like human robot interaction, human computer interaction and health-care system, 3D human body models and their dynamic motions has gained popularity. Human performance accompanies human body shapes and their relative motions. Research on human activity recognition is structured around how the complex movement of a human body is identified and analyzed. Vision based action recognition from video is such kind of tasks where actions are inferred by observing the complete set of action sequence performed by human. Many techniques have been revised over the recent decades in order to develop a robust as well as effective framework for action recognition. In this survey, we summarize recent advances in human action recognition, namely the machine learning approach, deep learning approach and evaluation of these approaches.

Manual interpretation of these huge amounts of image volumes are susceptible to inter-reader variability and human error. Thus, accurate automated CAD scheme is highly desirable in clinical pathological diagnosis. In this research, plethora of machine learning paradigms (e.g. feature extraction, dimensionality reduction and supervised classification methods) were explored, evaluated, compared and analyzed to identify the optimal pathway for brain MR images (normal vs neoplastic) binary classification task. External validation dataset was used to test the generalizability of the optimal predictive models implemented. Relevant and informative features were selected to construct cross-validated decision tree and eventually simple rule set was built based on the decision tree. The experimental results show that almost all pattern recognition paradigms achieve high accuracy with careful selection of number of attributes. LDA+ELM with 55 features are the optimal pipelines which achieve perfect classification when training and test data are of same source; and achieving (accuracy=97.5%, AUC=0.989, sensitivity=95% and specificity=100%) under balanced test dataset; (accuracy=99.5%, AUC=0.988, sensitivity=95% and specificity=100%). Cross-validated decision tree model also shows comparable performance: accuracy=98.8%, AUC=99.1%, sensitivity=99.6% and specificity=98.2%. Three highly relevant and robust attributes are visualized and selected for construction of decision tree models and finally a rule sets are read directly off the decision tree. This rule sets can potentially serve as fast and accurate classification algorithm.

This research investigates the potential benefits of integrating machine learning algorithms and IoT sensors in modern agriculture. The focus is on optimizing crop production and reducing waste through informed decisions about planting, watering, and harvesting crops. The paper discusses the current state of machine learning and IoT in agriculture, highlighting key challenges and opportunities. It also presents experimental results that demonstrate the impact of changing labels on the accuracy of data analysis algorithms. The findings recommend that by analyzing wide-ranging data collected from farms, including real-time data from IoT sensors, farmers can make more informed verdicts about factors that affect crop growth. Eventually, the integration of these technologies can transform modern agriculture by increasing crop yields while minimizing waste. In our studies, we achieve a classification accuracy of 99.59% using the Bayes Net algorithm and 99. 46% using Naïve Bayes Classifier, and Hoeffding Tree algorithms. Our results indicate that we achieved high accuracy results in our experiments in order to increase crop growth.

The rapid infiltration of fake news is a flaw to the otherwise valuable internet, a virtually global network that allows for the simultaneous exchange of information. While a common, and normally effective, approach to such classification tasks is designing a deep learning-based model, the subjectivity behind the writing and production of misleading news invalidates this technique. Deep learning models are unexplainable in nature, making the contextualization of results impossible because it lacks explicit features used in traditional machine learning. This paper emphasizes the need for feature engineering to effectively address this problem: containing the spread of fake news at the source, not after it has become globally prevalent. Insights from extracted features were used to manipulate the text, which was then tested on deep learning models. The original unknown yet substantial impact that the original features had on deep learning models was successfully depicted in this study.

Recently, spam emails have become a significant problem with the expanding usage of the Internet. It is to some extend obvious to filter emails. A spam filter is a system that detects undesired and malicious emails and blocks them from getting into the users' inboxes. Spam filters check emails for something "suspicious" in terms of text, email address, header, attachments, and language. However, we have used different features such as word2vec, word n-grams, character n-grams, and a combination of variable length n-grams for comparative analysis in our proposed approach. Different machine learning models such as support vector machine (SVM), decision tree (DT), logistic regression (LR), and multinomial naïve bayes (MNB) are applied to train the extracted features. We use different evaluation metrics such as precision, recall, f1-score, and accuracy to evaluate the experimental results. Among them, SVM provides 97.6 \% of accuracy, 98.8\% of precision, and 94.9\% of f1-score using a combination of n-gram features.

Solar radiation prediction is an important process in ensuring optimal exploitation of solar energy power. Numerous models have been applied to this problem, such as numerical weather prediction models and artificial intelligence models. However, well-designed hybridization approaches that combine numerical models with artificial intelligence models to yield a more powerful model can provide a significant improvement in prediction accuracy. In this paper, we propose novel hybrid machine learning approaches that exploit auxiliary numerical data. The proposed hybrid methods invoke different machine learning paradigms, including feature selection, classification, and regression. Additionally, numerical weather prediction (NWP) models are used in the proposed hybrid models. Feature selection is used for feature space dimension reduction to reduce the large number of recorded parameters that affect estimation and prediction processes. The rough set theory is applied for attribute reduction and the dependency degree is used as a fitness function. We investigate the effect of the attribute reduction process with thirty different classification and prediction models in addition to the proposed hybrid model. Then, different machine learning models are constructed based on classification and regression techniques to predict solar radiation. Moreover, other hybrid prediction models are formulated to use the output of the numerical model of Weather Research and Forecasting (WRF) as learning elements in order to improve the prediction accuracy. The proposed methodologies are evaluated using a data set that is collected from different regions in Saudi Arabia.

Deep learning (DL) models have been recently widely used to extract task-oriented patterns from large scale of datasets, and to improve the data image understanding and analysis accuracy in many different decision-making processes for tasks such as image classification, segmentation, detection, and so on. However, in practice, the performances of DL models are easily affected by environmental illumination conditions. Conversely, DL models can also be utilized for extracting the illumination hints from the images, and these hints are critically useful for improving the model robustness, classifying the environmental scenes, estimating scene depth information, and rendering 3D objects. In this study, an extensive and exhaustive review is carried out for DL based color constancy, indoor and outdoor illumination estimation, and image depth estimations with the considerations of strengths and weaknesses of DL models. This study also explores the different network designs and the paradoxes in parameter optimization during the model training. Current technology barriers involved in implementing these models and recommendations to overcome these barriers are also suggested in the review.

It is crucial to select the right machine learning classifier for image classification and face recog-nition. This study examines the effectiveness of four different face recognition classifiers - Support Vector Machines (SVM), Random Forest, K-Nearest Neighbors (KNN), and Neural Networks. An analysis of the Large Faces in the Wild (LFW) dataset was carried out using Principal Component Analysis (PCA). Classifiers are rigorously trained and evaluated based on the extracted features. Comparison of classifier performance is an insightful way to figure out their strengths and weaknesses. Having a visual representation of the classifier's performance gives a complete understanding of its capabilities. Through the selection of the most appropriate classifier, study results contribute to advancements in image classification, recognition, and biometric identification. The comparison study demonstrated that the Neural Network classifier was exceptionally accurate and proficient in recognizing faces from the LFW dataset when used in conjunction with PCA for feature extraction. According to the comparative analysis, the Neural Network classifier proved exceptionally accurate and proficient at identifying faces from the LFW dataset when combined with PCA for feature extraction.

AliAmbra is a project developed to explore and promote high-quality catches of the Amvrakikos Gulf GP to Artas’ wider regions. In addition, this project aimed to implement an integrated plan of action, to form a business identity with high-added value and achieve integrated business services adapted to the special characteristics of the area. The action plan for this project was to actively search for new markets, create a collective identity for the products, promote their quality and added value, engage in gastronomes and tasting exhibitions, dissemination and publicity actions, as well as enhance the quality of the products and markets based on the customer needs. The primary focus of this publication is to observe and analyze the data retrieved from various tasting exhibitions of the AliAmbra project, with a target goal of improving customer experience and product quality.

Structures inevitably suffer damages after an earthquake, with their severity ranging from minimal damages to non-structural elements to partial or even total collapse, possibly with loss of human lives. Thus, it is essential for engineers to understand the crucial factors that drive a structure towards suffering higher degrees of damage, so that preventative measures can be taken. In the present study, we focus on three well-known damage thresholds, namely Collapse Limit State, Ultimate Limit State and Serviceability Limit State and analyze which of the features obtained via Rapid Visual Screening, determine whether or not a given structure will cross that threshold. To this end, we use Machine Learning to perform binary classification for each damage threshold, as well as Explainability via SHAP values (acronym for SHapley Additive exPlanations) to quantify the effect of each parameter. The quantitative results obtained demonstrate the potential applicability of ML methods to re-calibrate the computation of structural vulnerability indices using data from recent earthquakes.

Seismic imaging is the main technology used for subsurface hydrocarbon prospection. It provides an image of the subsurface using the same principles as ultrasound medical imaging. It is based on emitting a sound (pressure) wave through the subsurface and recording the reflected echoes using hydrophones (pressure sensors) and/or geophones (velocity/acceleration sensors). Contrary to medical imaging, which is done in real time, subsurface seismic imaging is an offline process that involves a huge volume of data and needs considerable computing power. The raw seismic data are heavily contaminated with noise and unwanted reflections that need to be removed before further processing. Therefore, the noise attenuation is done at an early stage and often while acquiring the data. Quality control (QC) is mandatory to give confidence in the denoising process and to ensure that a costly data re-acquisition is not needed. QC is done manually by humans and comprises a major portion of the cost of a typical seismic processing project. It is therefore advantageous to automate this process to improve cost and efficiency. Here, we propose a supervised learning approach to build an automatic QC system. The QC system is an attribute-based classifier that is trained to classify three types of filtering (mild = underfiltering, noise remaining in the data; optimal = good filtering; harsh = overfiltering, the signal is distorted). The attributes are computed from the data and represent geophysical and statistical measures of the quality of the filtering. The system is tested on a full-scale survey (9000 km2) to QC the results of the swell noise attenuation process in marine seismic data. The results are encouraging and helped identify localized issues that were difficult for a human to spot.

Machine Learning techniques can be used to identify whether deficits in cognitive functions contribute to antisocial and aggressive behavior. This paper initially presents the results of tests conducted on delinquent and non-delinquent youths to assess their cognitive functions. The dataset extracted from these assessments, consisting of 37 predictor variables and one target. was used to train three algorithms that aim to predict whether the data corresponds to that of a young offender or a non-offending youth. Prior to this, statistical tests were conducted on the data to identify characteristics that exhibited significant differences in order to select the most relevant features and optimize the prediction results. Additionally, other feature selection methods, such as Boruta, RFE, and Filter, were applied, and their effects on the accuracy of each of the three machine learning models used (SVM, RF, and KNN) were compared. 80% of the data were utilized for training, while the remaining 20% were used for validation. The best result was achieved by the K-NN model trained with 19 features selected by the Boruta method, followed by the SVM model trained with 24 features selected by the filter method.

State-of-the-art remote sensing scene classification methods employ different Convolutional Neural Network architectures for achieving very high classification performance. A trait shared by the majority of these methods is that the class associated with each example is ascertained by examining the activations of the last fully connected layer, and the networks are trained to minimize the cross-entropy between predictions extracted from this layer and ground-truth annotations. In this work, we extend this paradigm by introducing an additional output branch which maps the inputs to low dimensional representations, effectively extracting additional feature representations of the inputs. The proposed model imposes additional distance constrains on these representations with respect to identified class representatives, in addition to the traditional categorical cross-entropy between predictions and ground-truth. By extending the typical cross-entropy loss function with a distance learning function, our proposed approach achieves significant gains across a wide set of benchmark datasets in terms of classification, while providing additional evidence related to class membership and classification confidence.

Background: Pancreatic cancer (PC) is a disease with poor prognosis and survival rate. There is a pertinent need to identify the risk factors of this disease. The purpose of this study is to identify a subset of factors (a.k.a. features) as predictors of PC from the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer dataset consisting of responses to 65 questions about demographics, cancer and health history, medication usage, and smoking habits from 154,897 participants. Method: There are two challenges to selecting the subset of features that predict PC with highest probability: the problem is computationally intractable, and the PLCO dataset is highly imbalanced. We use an innovative method to use the dataset in a balanced way, without involving up- or down-sampling. We use nine feature selection methods to select the optimal subset of features from the preprocessed and balanced dataset. Results: Our preprocessed dataset consists of 32 risk factors (8 demographics, 5 cancer history, 13 health history, 2 medication usage, 4 smoking habits). Risk factors belonging to cancer and health history, followed by smoking habits, were consistently chosen by the feature selection methods. We also discuss findings in the medical sciences literature that corroborate our findings. Conclusions: The study found that risk factors belonging to cancer and health history are the most prominent ones for PC. In particular, previously diagnosed with PC is chosen as the most prominent risk factor by majority of methods. While most of our findings are consistent with the literature, some of our findings shed light on novel factors that may not have received their due attention by the research community.

Bayesian formulations of learning imply that whenever the evidence for a correlation between events in an animal’s habitat is sufficient, the correlation is learned. This implies that regularities can be learnt rapidly, from small numbers of learning examples. This speed of learning gives maximum possible fitness, and no faster learning is possible. There is evidence in many domains that animals and people can learn at nearly Bayesian optimal speeds. These domains include associative conditioning, and the more complex domains of navigation and language. There are computational models of learning which learn at near-Bayesian speeds in complex domains, and which can scale well – to learn thousands of pieces of knowledge (i.e., relations and associations). These are not neural net models. They can be defined in computational terms, as algorithms and data structures at David Marr’s [1] Level Two. Their key data structures are composite feature structures, which are graphs of multiple linked nodes. This leads to the hypothesis that animal learning results not from deep neural nets (which typically require thousands of training exam-ples), but from neural implementations of the Level Two models of fast learning; and that neu-rons provide the facilities needed to implement those models at Marr’s Level Three. The required facilities include feature structures, dynamic binding, one-shot memory for many feature struc-tures, pattern-based associative retrieval, unification and generalization of feature structures. These may be supported by multiplexing of data and metadata in the same neural fibres.

This paper explores the use of different deep learning models for predicting precipitation in 56 meteorological stations in Jilin Province, China. The models used include Stacked-LSTM, Transformer, and SVR, and Gaussian noise is added to the data to improve their robustness. Results show that the Stacked-LSTM model performs the best, achieving high prediction accuracy and stability. The study also conducts variable attribution analysis using LightGBM and finds that temperature, dew point, precipitation in previous days, and air pressure are the most important factors affecting precipitation prediction, which is consistent with traditional meteorological theory. The paper provides detailed information on the data processing, model training, and parameter settings, which can serve as a reference for future precipitation prediction tasks. The findings suggest that adding Gaussian noise to the dataset can improve the model's generalization ability, especially for predicting days with zero precipitation. Overall, this study provides useful insights into the application of deep learning models in precipitation prediction and can contribute to the development of meteorological forecasting and applications.

In recent years, many algorithms based on end-to-end deep networks have been proposed to deal with the target detection problem of videos. However, the deep network models usually consume a lot of computing resources during the procedure of analysis of videos with complex dynamic backgrounds. In this paper, a new method of object detection based on information theory is presented. Firstly, each frame in a video is converted into an effective information map by using the Harris corner detection method. Secondly, the sensitive areas in the frame are extracted by using the context information and the effective information maps of the consecutive video frames. The sensitive areas in the video frame are the candidate areas where the target objects would be appeared at high probabilities. Thirdly, the information entropy features of each sensitive area are extracted to form the feature matrix, based on which, an SVM model is trained for selecting the target areas from the sensitive areas. Finally, the locations of the objects are detected based on the target areas in the video with a complex dynamic background. As a lightweight video detection framework, the method presented in this paper can save a lot of computing resources. Experimental results show that this method can achieve good results in the benchmark of CDnet 2014.

Stagnation at local optima represents a significant challenge in bio–inspired optimization algorithms, often leading to suboptimal solutions. This paper addresses this issue by proposing a hybrid model that combines the Orca Predator Algorithm with Deep Q–Learning. Orca Predator Algorithm is an optimization technique that mimics the hunting behavior of orcas. It solves complex optimization problems by exploring and exploiting search spaces efficiently. Deep Q–Learning is a reinforcement learning technique that combines Q–Learning with deep neural networks. This integration aims to turn the stagnation problem into an opportunity for more focused and effective exploitation, enhancing the optimization technique’s performance and accuracy. The proposed hybrid model leverages the biomimetic strengths of Orca Predator Algorithm to identify promising regions nearby in the search space, complemented by the fine–tuning capabilities of Deep Q–Learning to navigate these areas precisely. The practical application of this approach is evaluated using the high–dimensional Heartbeat Categorization Dataset, focusing on the feature selection problem. This dataset, comprising complex electrocardiogram signals, provided a robust platform for testing the feature selection capabilities of our hybrid model. Our experimental results are encouraging, showcasing the hybrid strategy capability to identify relevant features without significantly compromising the performance metrics of machine learning models. This analysis was performed by comparing the improved method of Orca Predator Algorithm against its native version and a set of state–of–the–art algorithms.

Epigenetic aging has been found associated with a number of phenotypes and diseases. Few studies investigated its effect on lung function in relatively older people. However, this effect has not been explored in younger population. This study examines whether lung function at adolescent can be predicted with epigenetic age accelerations (AAs) using machine learning techniques. DNA methylation based AAs were estimated in 326 matched samples at two time points (at 10 years and 18 years) from the Isle of Wight Birth Cohort. Five machine learning regression models (linear, lasso, ridge, elastic net, and Bayesian ridge) were used to predict FEV1 (Forced Expiratory Volume in one second) and FVC (Forced Vital Capacity) at 18 years from feature selected predictor variables (based on mutual information) and AA changes between the two time points. The best models were ridge regression (R2 = 75.21% ± 7.42%; RMSE = 0.3768 ± 0.0653) and elastic net regression (R2 = 75.38% ± 6.98%; RMSE = 0.445 ± 0.069) for FEV1 and FVC, respectively. This study suggests that the application of machine learning in conjunction with tracking changes in AA over life span can be beneficial to assess the lung health in adolescence.

In this study, machine learning was used to predict and analyze the behavior of occupants in Gifu City residences during winter. Global warming is currently progressing worldwide, and it is important to control greenhouse gas emissions from the perspective of adaptation and mitigation. Occupant behavior is highly individualized and must be analyzed to accurately determine a building's energy consumption. The accuracy of heating behavior prediction has been studied using three different methods: logistic regression, support vector machine (SVM), and deep neural network (DNN). The generalization ability of the support vector machine and the deep neural network was improved by parameter tuning. Parameter tuning of the SVM showed that the values of C and gamma affected the prediction accuracy. The prediction accuracy improved by approximately 11.9 %, confirming the effectiveness of parameter tuning on SVM. Parameter tuning of the DNN showed that the values of layer and neuron affected the prediction accuracy. Although parameter tuning also improved the prediction accuracy of DNN, and the rate of increase was lower than that of SVM.

Disease-related data and information collected by physicians, patients, and researchers seem insignificant at first glance. Still, the same unorganized data contain valuable information that is often hidden. The task of data mining techniques is to extract patterns to classify the data accurately. One of the various Data mining and its methods have been used often to diagnose various diseases. In this study, a machine learning (ML) technique based on distributed computing in the Apache Spark computing space is used to diagnose diabetics or hidden pattern of the illness to detect the disease using a large dataset in real-time. Implementation results of three ML techniques of Decision Tree (DT) technique or Random Forest (RF) or Support Vector Machine (SVM) in the Apache Spark computing environment using the Scala programming language and WEKA show that RF is more efficient and faster to diagnose diabetes in big data.

An autonomous concrete crack inspection system is necessary for preventing hazardous incidents arising from deteriorated concrete surfaces. In this paper, we represent a concrete crack detection framework to aid the process of automated inspection. The proposed approach employs a deep convolutional neural network architecture for crack segmentation from concrete image. The proposed network alleviates the effect of gradient vanishing problem present in deep neural network architectures. A feature silencing module is incorporated in the crack detection framework, for eliminating unnecessary feature maps from the network. The overall performance of the network significantly improves as a result. Experimental results support the benefit of incorporating feature silencing within a convolutional neural network architecture for improving the network’s robustness, sensitivity, and specificity. An added benefit of the proposed architecture is its ability to accommodate for the trade-off between specificity (positive class detection accuracy) and sensitivity (negative class detection accuracy) with respect to the target application. Furthermore, the proposed framework achieves a high precision rate and processing time than crack detection architectures present in literature.

Seismic imaging is the main technology used for subsurface hydrocarbon prospection. It~provides an image of the subsurface using the same principles as ultrasound medical imaging. As for any data acquired through hydrophones (pressure sensors) and/or geophones (velocity/acceleration sensors), the raw seismic data are heavily contaminated with noise and unwanted reflections that need to be removed before further processing. Therefore, the noise attenuation is done at an early stage and often while acquiring the data. Quality control (QC) is mandatory to give confidence in the denoising process and to ensure that a costly data re-acquisition is not needed. QC is done manually by humans and comprises a major portion of the cost of a typical seismic processing project. It is therefore advantageous to automate this process to improve cost and efficiency. Here, we propose a supervised learning approach to build an automatic QC system. The~QC system is an attribute-based classifier that is trained to classify three types of filtering (mild = under filtering, noise remaining in the data; optimal = good filtering; harsh = over filtering, the signal is distorted). The attributes are computed from the data and represent geophysical and statistical measures of the quality of the filtering. The system is tested on a full-scale survey (9000 km²) to QC the results of the swell noise attenuation process in marine seismic data.

Taste determination in small molecules is critical in food chemistry, but traditional experimental methods can be time-consuming. Consequently, computational techniques have emerged as val-uable tools for this task. In this study, we explore taste prediction using various molecular feature representations and assess the performance of different machine learning algorithms on a dataset comprising 2,601 molecules. The results reveal that GNN-based models outperform other ap-proaches in taste prediction. Moreover, consensus models that combine diverse molecular repre-sentations demonstrate improved performance. Among these, molecular fingerprints + GNN con-sensus model emerges as the top performer, highlighting the complementary strengths of GNNs and molecular fingerprints. These findings have significant implications for food chemistry research and related fields. By leveraging these computational approaches, taste prediction can be expedited, leading to advancements in understanding the relationship between molecular structure and taste perception in various food components and related compounds.

In South Korea, the residential real estate market is influenced not just by traditional supply and demand dynamics, but also by external factors such as housing policies and macroeconomic conditions. Given the significant role of housing assets in individual wealth, market fluctuations can have profound implications. While prior research has utilized variables like GDP growth rate, patent issuance, and birth rates, and employed models like LSTM and ARIMA for housing price predictions, many overlook key localized factors. Notably, the impact of subway stations and living SOC facilities on housing prices, especially in metropolitan areas, remains underexplored. This study addresses these gaps by analyzing usage trends across subway stations, assessing the influence of living SOC facilities on housing values, and identifying the optimal machine learning model for price predictions near transport hubs. Through a comparative analysis of machine learning techniques, we aim to provide insights for more informed housing price determinations, promoting a more stable real estate market.

Like traditional single label learning, multi-label learning is also faced with the problem of dimensional disaster.Feature selection is an effective technique for dimensionality reduction and learning efficiency improvement of high-dimensional data. In this paper, Logistic regression, manifold learning and sparse regularization were combined to construct a joint framework for multi-label feature selection (LMFS). Firstly, the sparsity of the eigenweight matrix is constrained by the $L_{2,1}$-norm. Secondly, the feature manifold and label manifold can constrain the feature weight matrix to make it fit the data information and label information better. An iterative updating algorithm is designed and the convergence of the algorithm is proved.Finally, the LMFS algorithm is compared with DRMFS, SCLS and other algorithms on eight classical multi-label data sets. The experimental results show the effectiveness of LMFS algorithm.

Due to the significant price difference, expensive oils like olive oil are often blended with cheaper edible oils. This practice of adulteration in edible oils, aimed at increasing profits for producers, poses a major concern for consumers. Furthermore, adulteration in edible oils can lead to various health issues impacting consumer well-being. In order to meet the requirements of fast, non-destructive, universal, accurate and reliable quality testing for edible oil, the oblique-incidence reflectivity difference (OIRD) method combined with machine learning algorithms was introduced to detect a variety of edible oils. The prediction accuracy of Gradient Boosting, K-Nearest Neighbor, and Random Forest models exceeded 95%. Experimental results indicate that the OIRD method can serve as a powerful tool for detecting edible oils.

The need for Artificial Intelligence (AI) and Machine Learning (ML) technologies is increasingly being leveraged for optimizing Data centers’ (DCs’) operations as the volume of operations management data increase tremendously. These strategies can assist operators in better understanding their DC operations and making informed decisions up front to preserve service reliability and availability. Aiming at creating models that optimize energy efficiency, identify inefficient resource utilization and scheduling policies, and predict outages. Apart from model hyperparameter tuning, feature selection is a crucial step to identify relevant features with the objective of providing insight into the data, improving performance, and reducing computational expenses. Although several feature selection methods have been discussed in various domains, none have been discussed in the context of the data center. This paper introduces SHapley Additive exPlanation (SHAP), a class of additive feature attribution values-based feature selection that is rarely discussed in literature. We compared the effectiveness of SHAP method with several widely used methods. We used a real DC dataset obtained from the ENEA CRESCO6 cluster with 2,0832 cores to evaluate the methods. To demonstrate the comparison of the methods, we picked the top 10 most important features from each method, the predictions were retrained, and their performance was evaluated using MAE, RMSE, and MPAE. The results show that the SHAP-assisted feature selection performed best and align with human intuition.

Microgrids are becoming popular nowadays because they provide clean, efficient, and low-cost energy. To use the stored energy in times of emergency or peak loads, microgrids require bulk storage capacity. Since microgrids are the future of renewable energy, the energy storage technology employed should be optimized to generate electricity. Batteries play a variety of essential roles in daily life and are used at peak hours and during a time of emergency. There are different types of batteries i.e., lion batteries, lead-acid batteries, etc. Optimal battery sizing of microgrids is a challenging problem, that limits modern technologies such as electric vehicles, etc. It is important to know different battery features such as battery life, battery throughput, and battery autonomy to get optimal battery sizing for microgrids. Mixed-integer linear programming (MILP) is an established technique for the integration and optimization of different energy sources and parameters for optimal battery sizing. A new MILP based dataset is introduced in this work. Support vector machine (SVM) is the machine learning application used to estimate the optimum battery size. The impact of feature selection algorithms on the proposed machine learning-based model is evaluated. The performance of the six best-performing feature selection algorithms is analyzed. The experimental results show that the feature selection algorithms improve the performance of the proposed methodology. Ranker search shows the best performance with a Spearman’s rank-ordered correlation constant of 0.9756, linear correlation constant of 0.9452, Kendall correlation constant of 0.8488 and root mean squared error of 0.0525.

Vibration monitoring is a critical aspect of assessing the health and performance of machinery and industrial processes. This study explores the application of machine learning techniques, specifically the Random Forest (RF) classification model, to predict and classify chatter—a detrimental self-excited vibration phenomenon—during machining operations. While sophisticated methods have been employed to address chatter, this research investigates the efficacy of a simplified RF model. The study leverages simulated vibration data, bypassing resource-intensive real-world data collection, to develop a versatile chatter detection model applicable across diverse machining configurations. The feature extraction process combines time-series features and Fast Fourier Transform (FFT) data features, streamlining the model while addressing challenges posed by feature selection. By focusing on the RF model’s simplicity and efficiency, this research advances chatter detection techniques, offering a practical tool with improved generalizability, computational efficiency, and ease of interpretation. The study demonstrates that innovation can reside in simplicity, opening avenues for wider applicability and accelerated progress in the machining industry.

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.

Due to the substantial growth of internet users and its spontaneous access via electronic devices, the amount of electronic contents is growing enormously in recent years through instant messaging, social networking posts, blogs, online portals, and other digital platforms. Unfortunately, the misapplication of technologies has boosted with this rapid growth of online content which leads to the rise in suspicious activities. People misuse the web media to disseminate malicious activity, perform the illegal movement, abuse other people, and publicize suspicious contents on the web. The suspicious contents usually available in the form of text, audio or video, whereas text contents have been used in most of the cases to perform suspicious activities. Thus, one of the most challenging issues for NLP researchers is to develop a system that can identify suspicious text efficiently from the specific contents. In this paper, a Machine Learning (ML)-based classification model is proposed (hereafter called STD) to classify Bengali text into non-suspicious and suspicious categories based on its original contents. A set of ML classifiers with various features has been used on our developed corpus, consisting of 7000 Bengali text documents where 5600 documents used for training and 1400 documents used for testing. The performance of the proposed system is compared with the human baseline and existing ML techniques. The SGD classifier `tf-idf’ with the combination of unigram and bigram features are used to achieve the highest accuracy of 84.57%.

Genomics involving tens of thousands of genes is a complex system determining phenotype. An interesting and vital issue is that how to integrate highly sparse genetic genomics data with a mass of minor effects into prediction model for improving prediction power. We find that deep learning method can work well to extract features by transforming highly sparse dichotomous data to lower dimensional continuous data in a non-linear way. This idea may provide benefits in risk prediction based on genome-wide data associated e.g. integrating most of the information in the genotype data. Hence, we developed a multi-stage strategy to extract information from highly sparse binary genotype data and applied it for risk prediction. Specifically, we first reduced the number of biomarkers via a univariable regression model to a moderate size. Then a trainable auto-encoder was used to extract compact representations from the reduced data. Next, we performed a LASSO problem process over a grid of tuning parameter values to select the optimal combination of extracted features. Finally, we applied such feature combination to two prognostic models, and evaluated predictive effect of the models. The results of simulation studies and real data applying indicated that these highly compressed transformation features could better improve predictive performance and did not easily lead to over-fitting.

Despite being considered as natural components of many ecosystems, forest fires pose significant threats to the environment and human health. In order to ensure public safety and effective fire suppression planning, it is necessary to develop reliable prediction models to mitigate forest fire danger. These models should account for specific environmental conditions. The advent of big data in recent years has opened new avenues for improving forest fire predictions. Machine learning techniques, surpassing traditional forecasting methods, have shown significant promise in this area. By applying the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology to a real-world dataset, this paper explores the application of machine learning approaches to understand forest fire patterns and predict fire danger. We consider six distinct time stages and incorporate feature selection to refine our predictions. It is important to note that forest fire behavior models are not universally effective due to geographical variations in data. Nevertheless, advanced decision-making techniques are vital in forest fire management. Our research presents a systematic exploration of the topic, comparing various machine learning models, thereby providing a comprehensive baseline for future investigations in this crucial environmental arena.

COVID-19 vaccine hesitancy is considered responsible for the lower rate of acceptance of vaccines in many parts of the world. However, sources of this hesitancy are rooted in many social, political, and economic factors. This paper strives to find the most important variables in predicting the COVID-19 vaccination uptake. We introduce an explainable machine learning (ML) framework to understand the COVID-19 vaccination uptake around the world. To predict vaccination uptake, we have trained a random forest (RF) regression model using a number of sociodemographic and socioeconomic data. The traditional decision tree (DT) regression model is also implemented as the baseline model. We found that the RF model performed better than the DT model since RF is more robust to handle nonlinearity and multi-collinearity. Also, we have presented feature importance based on impurity measure, permutation, and Shapley values to provide the most significant unbiased features. It is found that electrification coverage and Gross Domestic Product are the strongest predictors for higher vaccination uptake, whereas the Fragile state index (FI) contributed to lower vaccination uptake. These findings suggest addressing issues that are found responsible for lower vaccination uptake to combat any future public health crisis.

Deep learning is based on matrix computing with a large amount of hidden parameters that is not visible outside the computing module. Deep learning is a non-linear system and a linear approach is not possible. It is natural for people to visualize the algorithm and to follow some hidden parameters. In this paper, we propose a simple graphical programming of a nonlinear system based on drawing the simplest unit, such as a single neuron model. A more complex scheme of a classical neural network is obtained by commands copy-move-place of one neuron. The number of neurons and layers can be chosen arbitrarily. Once the scheme is complete, the implementation code is evaluated with symbolic parameters and nonlinear activation functions. This cannot be done manually. With the symbolic expression of outputs in terms of inputs and symbolic parameters, including symbolic activation pure functions, some other properties can be derived in closed form. This unique original approach can help scientists and developers and design numerical algorithms for machine learning and to understand how deep learning algorithms work.

Few-shot fine-grained image classification (FSFGIC) methods refer to machine learning methods which aim to classify images (e.g., bird species, flowers, and airplanes) belonging to subordinate object categories of the same entry-level category with only a few samples. It is worth to note that feature representation learning is used not only to represent training samples, but also to construct classifiers for performing various FSFGIC tasks. In this paper, starting from the definition of FSFGIC, a taxonomy of feature representation learning for FSFGIC is proposed. According to this taxonomy, we discuss key issues on FSFGIC (including data augmentation, local or/and global deep feature representation learning, class representation learning, and task specific feature representation learning). The existing popular datasets and evaluation standards are introduced. Furthermore, a novel classification performance evaluation mechanism is designed with a 0.95 confidence interval for judging whether the classification accuracy obtained by a certain specified method is good or bad. Moreover, current challenges and future trends of feature representation learning on FSFGIC are elaborated.

The used car market has a high global economic importance, with more than 35 million cars sold yearly. Accurately predicting prices is a crucial task for both buyers and sellers to facilitate informed decisions in terms of opportunities or potential problems. Although various Machine Learning techniques have been applied to create robust prediction models, a comprehensive approach has yet to be studied. This research introduces two datasets from different markets, one with over 300,000 entries from Germany to serve as a training base for deep prediction models and a second dataset from Romania containing more than 15,000 car quotes used mainly to observe local traits. As such, we include extensive cross-market analyses by comparing the emerging Romanian market versus one of the world’s largest and most developed car markets, Germany. Our study uses several neural network architectures that capture complex relationships between car model features, individual add-ons, and visual features to predict used car prices accurately. Our models achieved a high R2 score exceeding 0.95 on both datasets, indicating their effectiveness in estimating used car prices. Moreover, we experimented with advanced convolutional architectures to predict car prices based solely on visual features extracted from car images. This approach exhibited transfer-learning capabilities, leading to improved prediction accuracy, especially since the Romanian training dataset is limited. Our experiments highlight the most important factors influencing the price, while our findings have practical implications for buyers and sellers in assessing the value of vehicles. At the same time, the insights gained from this study enable informed decision-making and provide valuable guidance in the used car market.

The present paper explored the possibility of discriminating between different malignant degrees of Chondrosarcoma (CS), a diffused bone tumor, and Enchondroma (EC), its benign version, through an approach based on the coupling between Confocal Raman Microscopy (CRM) and Machine Learning (ML) techniques. Recently, Raman Spectroscopy has proven to be a powerful tool for a complete grading of CS by distinguishing between grade 1, 2 and 3, and to distinguish CS from EC. In this paper, we tested some models, belonging either to ML or to the sub-type called Deep Learning (DL), showing excellent classification performances, especially the DL algorithms, with classification accuracy approaching to 100%, making the models promising for future implementations of Raman spectroscopy when applied to oncology diagnosis. In turn, we highlighted how some proper ML models result in worse classification performances but better resolution of specific chemical target compounds, possible candidate to become malignant markers, with relevant implication for a correct diagnosis.

Accurate estimation of body fat percentage is essential for various health and fitness applications. Traditional methods for measuring body fat, such as underwater weighing, can be costly and inconvenient. In this study, we apply machine learning techniques to predict body fat percentage using easily accessible body measurements and data. The results highlight the effectiveness of regression models in estimating body fat, with Linear Regression, Ridge Regression, and Bayesian Ridge models achieving R-squared scores of approximately 74%, 73%, and 73%, respectively. These models provide a practical and cost-effective solution for individuals and professionals seeking reliable body fat estimates.

Nutrient regulation in aquaponic environments has been the topic of research for many years. Most have focused on appropriate control of nutrients in an aquaponic set-up, but very little research has been done on commercial scale applications. In our model, the input data was sourced on a weekly basis from three commercial aquaponic farms in South-East Texas over the course of a year. Due to limited number of data points, dimensionality reduction techniques like pair-wise correlation matrix was used to remove the highly correlated predictors. Feature selection techniques like the XGBoost classifier and Recursive Feature Elimination with ExtraTreesClassifier were used to rank the features in order of their relative importance. Ammonium and calcium were found to be the top two nutrient predictors and based on the months in which lettuce was cultivated, the median of these nutrient values from the historical dataset served as the optimal concentrations to be maintained in the aquaponic solution. To accomplish this, Vernier sensors were used to measure the nutrient values and actuator systems were built to dispense the appropriate nutrient into the ecosystem via a closed loop.

Internet users have significantly increased as a result of the spread of Internet of Things (IoT) technologies and 5G networks. But these developments also make people more susceptible to cybercrime. Intrusion detection systems (IDSs), which protect against cyber threats and facilitate early response, have emerged as crucial security measures to handle this expanding risk. This study intends to present a comprehensive review of IDS, how it interacts with machine learning (ML), and develop a suitable approach for attack detection in 5G and IoT environments. To accomplish this, we leverage the AWID dataset, which is the first wireless traffic dataset specifically designed for security purposes, focusing on the IEEE 802.11 standard and developed to the AWID3 dataset. In this research, we suggest a powerful machine-learning framework for wireless system intrusion detection. We perform evaluations in three stages, covering scenarios for multiple nominal classes, multiple numeric classes, and binary classes. In order to improve the performance of the intrusion detection model, we also use feature selection approaches. Additionally, we offer a model that incorporates the outcomes of three feature selection techniques, highlighting how crucial it is to comprehend the features present in wireless datasets. Our experiments demonstrate how a machine learning-based approach can detect attacks with a high level of accuracy. In particular, the boosted decision tree performs best when overlapping feature selection procedures, whereas the Logistic Regression approach obtains the maximum accuracy of 99% in the first two phases. By providing a comprehensive framework for identifying attacks in 5G and IoT contexts using machine learning approaches, this research makes a contribution to the field of intrusion detection. The results underline how important it is to comprehend wireless dataset characteristics and highlight the possibility of ML-based methods for attaining highly accurate intrusion detection.

Accurate weight measurement is pivotal for monitoring the growth and well-being of cattle. However, the conventional weighing process, which involves physically placing cattle on scales, is labor-intensive and distressing for the animals. Hence, the development of automated cattle weight prediction techniques assumes critical significance. This study proposes a weight prediction approach for Korean cattle using 3D segmentation-based feature extraction and regression machine learning techniques from incomplete 3D shapes acquired from real farm environments. In the initial phase, we generated mesh data of 3D Korean cattle shapes using a multiple-camera system. Subsequently, deep learning-based 3D segmentation with the PointNet network model was employed to segment two dominant parts of the cattle. From these segmented parts, three crucial dimensions of Korean cattle were extracted. Finally, we implemented five regression machine learning models (CatBoost regression, LightGBM, Polynomial regression, Random Forest regression, and XGBoost regression) for weight prediction. To validate our approach, we captured 270 Korean cattle in various poses, totaling 1190 poses of 270 cattle. The best result was achieved with mean absolute error (MAE) of 25.2 kg and mean absolute percent error (MAPE) of 5.81% using the random forest regression model.

Generalized zero-shot learning (GZSL) aims to simultaneously recognize both seen classes and unseen classes by training only on seen class samples and auxiliary semantic descriptions. Recent state-of-the-art methods infer unseen classes based on semantic information or synthesize unseen classes using generative models based on semantic information, all of which rely on the correct alignment of visual-semantic features. However, they often overlook the inconsistency between original visual features and semantic attributes. Additionally, due to the existence of cross-modal dataset biases, the visual features extracted and synthesized by the model may also mismatch with some semantic features, which could hinder the model from properly aligning visual-semantic features. To address this issue, this paper proposes a GZSL framework that enhances the consistency of visual-semantic features using self-distillation and disentanglement network (SDDN). The aim is to utilize self-distillation and disentanglement network to obtain semantically consistent refined visual features and non-redundant semantic features to enhance the consistency of visual-semantic features. Firstly, SDDN utilizes self-distillation technology to refine the extracted and synthesized visual features of the model. Subsequently, the visual-semantic features are then disentangled and aligned using a disentanglement network to enhance the consistency of the visual-semantic features. Finally, the consistent visual-semantic features are fused to jointly train a GZSL classifier. Extensive experiments demonstrate that the proposed method achieves more competitive results on four challenging benchmark datasets (AWA2, CUB, FLO, and SUN).

Cross flow heat exchangers are commonly used in the thermal industry to transfer heat from hot tubes to cooling fluid. To protect the heat exchanger tubes from corrosion and dust accumulation, microscale coatings are often applied. In this study, we present machine-learning models for predicting heat transfer from hot tubes with different micro-sized coatings to cooling fluid in a turbulent flow using computational fluid dynamics simulations. A dataset of approximately 1000 cases was generated by varying the coating coverage thickness of each tube, the inlet Reynolds number, fluid flow inlet temperature, and wall temperature of tubes. The machine-learning models were generated to predict the overall heat flow rate in the heat exchanger, and it was found that combining the features based on their importance preserved the accuracy of the models while maintaining all the relevant information. The simulation results demonstrate that the proposed method increases the coefficient of determination (R2) for the models. The R2 values for unseen data for Random Forest, K-Nearest Neighbors, and Support Vector Regression were 0.9810, 0.9037, and 0.9754, respectively, indicating the usefulness of the proposed model for predicting heat transfer in various types of heat exchangers.

Breast cancer is the most common malignancy in women worldwide. The pathogenesis of this disease is closely related to the estrogen receptor alpha subtype (ERα). Therefore, it is of great importance to develop effective inhibitors of ERα activity for the treatment of breast cancer. In this paper, we propose a novel ensemble machine learning model for quantitative structure-activity relationship of anti-breast cancer drugs, which can effectively predict drug activity in small samples with multiple characteristic variables. To avoid the problem of over-fitting caused by low-correlation independent variables, the scoring mechanism of random forest was improved by incorporating three relevance indicators, including the maximum mutual information number, Pearson correlation coefficient and distance correlation coefficient, and 20 optimal molecular descriptors were selected. The Bayesian hyperparameter optimization method was used to optimize the parameters of multiple linear regression (MLR), support vector regression (SVR), and extreme gradient boosting (XGBoost), respectively. The AdaBoost strong learner was constructed by combining the weak learner with the weighted linear addition method. The results show that the proposed ensemble learning model has the best prediction performance compared to the three basic learner models and the CNN-LSTM combination prediction model. The root mean square error was reduced by 7.60%-26.51%. The mean relative error was reduced by 6.46%-30.92%. Goodness of fit increased by 9.57%-36.94%. Finally, the biological activities of 50 candidate compounds for ERα inhibitors were predicted, and it was found that 4-[2-benzyl-1-[4-(2-pyrrolidin-1-ylethoxy)phenyl]but-1-enyl]phenol had an excellent biological activity value pIC50, which had the potential to be an ERα inhibitor. The model proposed in this paper has good prediction accuracy, which can provide an effective reference for the discovery and development of anti-breast cancer drugs.

This paper offers a comprehensive examination of the process involved in developing and automating supervised end-to-end machine learning workflows for forecasting and classification purposes. It offers a complete overview of the components (i.e. feature engineering, model selection, etc), principles (i.e. bias-variance decomposition, model complexity, overfitting, model sensitivity to feature assumptions and scaling, output interpretability, etc), models (i.e. neural networks, regression models, etc), methods (i.e. Cross-Validation, data augmentation, etc), metrics (i.e. Mean Squared Error, F1-score, etc) and tools that rule most supervised learning applications with numerical and categorical data, as well as their integration, automation, and deployment. The end goal and contribution of this paper is the education and guidance of the non-AI expert academic community over complete and rigorous machine learning pipelines and data science practices, from problem scoping to design and state-of-the-art automation tools, including basic principles and reasoning in the choice of methods. The paper delves into the critical stages of supervised machine learning workflow development, many of which are often omitted by researchers due to brevity, and covers foundational concepts essential for understanding and optimizing a functional machine learning workflow, thereby offering a holistic view of task-specific application development for applied researchers who are not AI experts.

. 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.

Due to the intermittency and fluctuation of photovoltaic (PV) output power, a high proportion of grid-connected PV power generation systems has a significant impact on power systems. Accurate PV power forecasting can alleviate the uncertainty of the PV power and is of great significance for the stable operation and scheduling of the power systems. Therefore, in this study, a feature rise-dimensional (FRD) two-layer ensemble learning (TLEL) model for short-term PV power deterministic forecasting and probability forecasting is proposed. First， based on the eXtreme Gradient Boosting (XGBoost), Random Forest (RF), CatBoost, and Long-short-term memory (LSTM) models, a TLEL model is constructed utilizing the ensemble learning algorithm. Meanwhile, the FRD method is introduced to construct the FRD-XGBoost-LSTM (R-XGBL), FRD-RF-LSTM(R-RFL), and FRD- CatBoost - LSTM (R-CatBL) models. Subsequently, the above models are combined to construct the FRD-TLEL model for deterministic forecasting, and perform probability interval forecasting based on quantile regression(QR). Finally, the performance of the proposed model is demonstrated with a real-world dataset. By comparing with other models, the proposed model displays better forecasting accuracy for deterministic forecasting and reliable forecasting intervals for probability forecasting, and good generalization ability in the datasets of different seasons and weather types.

Heart disease is a global health concern of paramount importance, causing a significant number of fatalities and disabilities. Precise and timely diagnosis of heart disease is pivotal in pre-venting adverse outcomes and improving patient well-being, thereby creating a growing demand for intelligent approaches to predict heart disease effectively. This paper introduces an Ensemble Heuristic-Metaheuristic Feature Fusion Learning (EHMFFL) algorithm for heart disease diagnosis. Within the EHMFFL algorithm, a diverse ensemble learning model is crafted, featuring different feature subsets for each heterogeneous base learner, including support vector machine, K-nearest neighbors, logistic regression, random forest, naive bayes, decision tree, and XGBoost. The primary objective is to identify the most pertinent features for each base learner, leveraging a combined heuristic-metaheuristic approach that integrates the heuristic knowledge of Pearson correlation coefficient with the metaheuristic-driven grey wolf optimizer. The second objective is to aggregate the decision outcomes of the various base learners through ensemble learning, aimed at constructing a robust prediction model. The performance of the EHMFFL algorithm is rigorously assessed using the Cleveland and Statlog datasets yielding remarkable results with an accuracy of 91.8% and 88.9%, respectively, surpassing state-of-the-art machine learning, ensemble learning, and feature selection techniques in heart disease diagnosis. These findings underscore the potential of the EHMFFL algorithm in enhancing diagnostic accuracy for heart disease and providing valuable support to clinicians in making more informed decisions regarding patient care.

One of the most common and important destructive attacks on the victim system is Advanced Persistent Threats (APT)-attack. The APT attacker can achieve his hostile goals by obtaining information and gaining financial benefits regarding the infrastructure of a network. One of the solutions to detect a secret APT attack is using network traffic. Due to the nature of the APT attack in terms of being on the network for a long time and the fact that the network may crash because of high traffic, it is difficult to detect this type of attack. Hence, in this study, machine learning methods such as C5.0 decision tree, Bayesian network and deep neural network are used for timely detection and classification of APT-attacks on the NSL-KDD data set. Moreover, 10-fold cross validation method is used to experiment these models. As a result, the accuracy (ACC) of the C5.0 decision tree, Bayesian network and 6-layer deep learning models is obtained as 95.64%, 88.37% and 98.85%, respectively, and also, in terms of the important criterion of the false positive rate (FPR), the FPR value for the C5.0 decision tree, Bayesian network and 6-layer deep learning models is obtained as 2.56, 10.47 and 1.13, respectively. Other criterions such as sensitivity, specificity, accuracy, false negative rate and F-measure are also investigated for the models, and the experimental results show that the deep learning model with automatic multi-layered extraction of features has the best performance for timely detection of an APT-attack comparing to other classification models.

An effective process monitoring strategy is a requirement for meeting the challenges posed by increasingly complex products and manufacturing processes. To address these needs, this study investigates a comprehensive scheme based on classical machine learning methods, deep learning algorithms, and feature extraction and selection techniques. In a first step, a novel deep learning architecture based on convolutional neural networks (CNN) and gated recurrent units (GRU) is introduced to predict the local weld quality based on mid-wave infrared (MWIR) and near-infrared (NIR) image data. The developed technology is used to discover critical welding defects including lack of fusion (false friends), sagging and lack of penetration, and geometric deviations of the weld seam. Additional work is conducted to investigate the significance of various geometrical, statistical, and spatio-temporal features extracted from the keyhole and weld pool regions. Furthermore, the performance of the proposed deep learning architecture is compared to that of classical supervised machine learning algorithms, such as multi-layer perceptron (MLP), logistic regression (LogReg), support vector machines (SVM), decision trees (DT), random forest (RF) and k-Nearest Neighbors (kNN). Optimal hyperparameters for each algorithm are determined by an extensive grid search. Ultimately, the three best classification models are combined into an ensemble classifier that yields the highest detection rates and achieves the most robust estimation of welding defects among all classifiers studied, which is validated on previously unknown welding trials.

The presented research faces the problem of early detection of sepsis for patients in the Intensive Care Unit. The PhysioNet/Computing in Cardiology Challenge 2019 facilitated the development of automated, open-source algorithms for the early detection of sepsis from clinical data. A labeled clinical records dataset for training and verification of the algorithms was provided by the challenge organizers. However, a relatively small number of records with sepsis, supported by Sepsis-3 clinical criteria, led to highly unbalanced dataset (only 2% records with sepsis label). A high number of unbalanced data records is a great challenge for machine learning model training and is not suitable for training classical classifiers. To address these issues, a number of various models were investigated. A solution including feature selection and data balancing techniques was proposed in this paper. In addition, several performance metrics were investigated. Results show, that for successful prediction, a particular model having few or more predictors based on the length of stay in the Intensive Care Unit should be applied.

Face swapping is an intriguing and intricate task in the field of computer vision. Currently, most mainstream face swapping methods employ face recognition models to extract identity features and inject them into the generation process. Nonetheless, such methods often struggle to effectively transfer identity information, result in generated results failing to achieve a high identity similarity with the source face. Furthermore, if we can accurately disentangle identity information, we can achieve controllable face swapping, thereby providing more choices to users. In pursuit of this goal, we propose a new face swapping framework (ControlFace) based on the disentanglement of identity information. We disentangle the structure and texture of the source face, encoding and characterizing them in the form of feature embeddings separately. According to the semantic level of each feature representation, we inject them into the corresponding feature mapper and fuse them adequately in the latent space of StyleGAN. Owing to such disentanglement of structure and texture, we are able to controllablely transfer parts of the identity features. Extensive experiments and comparisons with state-of-the-art face swapping methods demonstrate the superiority of our face swapping framework in terms of transferring identity information, producing high-quality face images and controllable face swapping.

The central goal of few-shot scene classification is to learn a model that can generalize well to a novel scene category (UNSEEN) from only one or a few labeled examples. Recent works in the remote sensing (RS) community tackle this challenge by developing algorithms in a meta-learning manner. However, most prior approaches have either focused on rapidly optimizing a meta-learner or aimed at finding good similarity metrics while overlooking the embedding power. Here we propose a novel Task-Adaptive Embedding Learning (TAEL) framework that complements the existing methods by giving full play to feature embedding’s dual roles in few-shot scene classification - representing images and constructing classifiers in the embedding space. First, we design a lightweight network that enriches the diversity and expressive capacity of embeddings by dynamically fusing information from multiple kernels. Second, we present a task-adaptive strategy that helps to generate more discriminative representations by transforming the universal embeddings into task-specific embeddings via a self-attention mechanism. We evaluate our model in the standard few-shot learning setting on two challenging datasets: NWPU-RESISC4 and RSD46-WHU. Experimental results demonstrate that, on all tasks, our method achieves state-of-the-art performance by a significant margin.

Hybrid forecasting methods have emerged as a solution surpassing the limitations of both statistical and deep learning approaches. While the first emphasize the significance of variables, they often produce worse forecasting results when compared to newer techniques. In contrast, deep learning models remain enigmatic "black boxes" in terms of interpretability, although achieving better results in forecasting. This article introduces the Comprehensive Cross-Correlation and Lagged Linear Regression Deep Learning (CCLR-DL) framework, designed to harness the best of both approaches, enhancing forecasting accuracy while retaining model interpretability through a feature selection process. CCLR-DL blens cross-correlation analysis, lagged multiple linear regression and granger's causality procedures with deep learning architectures based on LSTM. In a practical demonstration, CCLR-DL was applied to a real database of clinical visits associated to diagnoses in Catalonia, Spain (tracking a population of 6.3 million patients during 10 years). Predicting visits enables the healthcare managers to be ready for future demand shifts. Results demonstrate a consistent and substantial improvement over standalone statistical and deep learning methods when predicting healthcare demand. This hybrid approach not only showcases its efficacy but also offers a promising solution to the challenge of balancing predictive accuracy with model explicability. In this context, this work aims to design and validate a method for feature selection and forecasting of multivariate high dimensional time series datasets not only to improve prediction accuracy but also to model transparency by identifying a subset of variables that improve predictions and G-cause the target variable.

In the last two decades, there have been massive advancements in high throughput technologies, which resulted in the exponential growth of public repositories of gene expression datasets for various phenotypes. It is possible to unravel biomarkers by comparing the gene expression levels under different conditions, such as disease vs. control, treated vs. not treated, drug A vs. drug B, etc. This problem refers to a well-studied problem in the machine learning domain, i.e., the feature selection problem. In biological data analysis, most of the computational feature selection methodologies were taken from other fields, without considering the nature of the biological data. For gene expression data analysis, most of the existing feature selection methods rely on expression values alone to select the genes; and biological knowledge is integrated at the end of the analysis in order to gain biological insights or to support the initial findings. Thus, integrative approaches that utilize the biological knowledge while performing feature selection are necessary for this kind of data. The main idea behind the integrative gene selection process is to generate a ranked list of genes considering both the statistical metrics that are applied to the gene expression data, and the biological background information which is provided as external datasets. Since the integrative approach attracted attention in the gene expression domain, lately the gene selection process shifted from being purely data-centric to more incorporative analysis with additional biological knowledge.

Breast cancer is a significant health issue across the world. Breast cancer is the most widely-diagnosed cancer in women; early-stage diagnosis of disease and therapies increase patient safety. This paper proposes a synthetic model set of features focused on the optimization of the genetic algorithm (CHFS-BOGA) to forecast breast cancer. This hybrid feature selection approach combines the advantages of three filter feature selection approaches with an optimize Genetic Algorithm (OGA) to select the best features to improve the performance of the classification process and scalability. We propose OGA by improving the initial population generating and genetic operators using the results of filter approaches as some prior information with using the C4.5 decision tree classifier as a fitness function instead of probability and random selection. The authors collected available updated data from Wisconsin UCI machine learning with a total of 569 rows and 32 columns. The dataset evaluated using an explorer set of weka data mining open-source software for the analysis purpose. The results show that the proposed hybrid feature selection approach significantly outperforms the single filter approaches and principal component analysis (PCA) for optimum feature selection. These characteristics are good indicators for the return prediction. The highest accuracy achieved with the proposed system before (CHFS-BOGA) using the support vector machine (SVM) classifiers was 97.3%. The highest accuracy after (CHFS-BOGA-SVM) was 98.25% on split 70.0% train, remainder test, and 100% on the full training set. Moreover, the receiver operating characteristic (ROC) curve was equal to 1.0. The results showed that the proposed (CHFS-BOGA-SVM) system was able to accurately classify the type of breast tumor, whether malignant or benign.

With the rapid advancements in remote sensing technology, the spectral information from hyperspectral remote sensing images has become increasingly rich, facilitating detailed spectral analysis of the Earth's surface objects. However, this abundance of spectral data poses significant challenges in data processing, such as the curse of dimensionality leading to the “Hughes” phenomenon, "strong correlation" due to high resolution, and "non-linear characteristics" caused by varied surface reflectance rates. Therefore, dimensionality reduction of hyperspectral data has become a crucial task. This paper, grounded in manifold theory and manifold learning techniques, and considering the non-linear structures and features in hyperspectral remote sensing data, elucidates the principles and processes of dimensionality reduction in hyperspectral remote sensing images using manifold learning, with a formalized expression of the process. This article introduces spectral information divergence (SID) into the nearest neighbor graph computation of manifold learning algorithms. The principles and computational processes of nearest neighbor graph algorithms based on Euclidean distance (ED), spectral angle mapping (SAM), and SID are studied, and a comparative analysis of the dimensionality reduction effects under these three metrics in hyperspectral data is conducted. Experiments on feature extraction under different metrics were performed using the publicly available Indian Pines hyperspectral dataset. The intrinsic features obtained post-dimensionality reduction were used as inputs for classification algorithms in ground objects classification experiments, with algorithm runtime, overall accuracy, and Kappa coefficient as evaluation metrics for dimensionality reduction quality. The results demonstrate that nearest neighbor graph computation based on SAM and SID outperforms traditional ED methods; SAM-based computation has the lowest time complexity, while SID-based manifold learning yields the highest accuracy in ground objects classification. Thus, manifold learning based on SAM and SID metrics proves to be an effective method for feature extraction in hyperspectral remote sensing data, underscoring the potential of manifold learning techniques in the dimensionality reduction of hyperspectral remote sensing images.

A Teoria das diferenças entre elementos explica as razões das diferenças nos conceitos físicos entre diferentes elementos, como a Física Quântica e a Teoria da Relatividade Geral, sendo explicada através de observações. Assim, é possível concluir a partir das observações que as diferenças são motivadas pelas reações entre os elementos, pela composição dos elementos, e pela influência dos espaços que promovem perdas e ganhos de características; portanto, não há semelhança geral entre a física quântica e a teoria geral da relatividade. O estudo tem como objetivo compreender os elementos e seus conceitos físicos que compõem o Universo.

Wearing gloves while operating machinery in workshops is an essential precaution to prevent mechanical injuries and burns from high temperatures, among other potential hazards. Ensuring workers are properly equipped with gloves, which is a crucial measure in accident prevention. Glove images often occupy a minimal proportion of the frame and are easily obscured by cluttered backgrounds, especially with limited edge computing resources. Consequently, this study proposes a glove detection algorithm called YOLOv8-AFPN-M-C2f based on YOLOv8, offering swifter detection speeds, lower computational demands, and enhanced accuracy for workshop scenarios. This research innovates by substituting the head of YOLOv8 with the AFPN-M-C2f network, amplifying the pathways for feature vector propagation, and mitigating semantic discrepancies between non-adjacent feature layers. Additionally, the introduction of a superficial feature layer enriches surface feature information, augmenting the model's sensitivity to smaller objects. To assess the performance of the YOLOv8-AFPN-M-C2f model, we conducted multiple experiments using a factory glove detection dataset compiled for this study. The results indicate that the enhanced YOLOv8 model surpasses other network models. Compared to the baseline YOLOv8 model, the refined version shows a 2.6% increase in mAP@50%, a 90.1% rise in FPS, and a 13% reduction in the number of parameters. This research contributes an effective solution for the detection of glove adherence.

Panoramic and periapical radiograph tools help dentists diagnose the most common dental diseases. Generally, dentists identify dental caries manually by inspecting X-ray images. However, due to their heavy workload, or poor image quality, dentists may sometimes overlook some unnoticeable dental caries, which may ultimately hinder the patient treatment. The purpose of this study was to develop an algorithm that classifies the teeth X-Ray images into three categories of “Normal”, “Caries”, and “Filled”. Our study used a dataset of 116 patients and 3712 single teeth images for training, validation and testing. Images were pre-processed using a sharpening filter and an intensity color map. We used a pre-trained transfer learning model, the NASNetMobile, which served as the feature extractor and the Convolutional Neural Network (CNN) model served as the classifier. The training dataset had a “Recall” of 0.92, 0.90 and 0.91 for “Normal”, “Caries” and “Filled” respectively and the test dataset had a “Recall” of 0.86, 0.81 and 0.85 for “Normal”, “Caries” and “Filled” respectively. The classification of the teeth X-Rays was successful and can be valuable for dentist as the artificial intelligence algorithm can serve as a decision support tool to aid dentists when they need to diagnose dental treatment.

Abstract: In recent years, the construction engineering sector has undergone a transformative shift towards the integration of machine learning (ML) techniques, particularly in predicting the properties of steel fiber-reinforced concrete (SFRC). Despite the theoretical sophistication of existing models, a persistent challenge remains – their opacity, lacking transparency and real-world relevance for practi-tioners. To address this gap and advance our current understanding, this study employs the extra gradient (XG) Boosting algorithm, crafting a comprehensive ap-proach. Grounded in a meticulously curated database drawn from 43 seminal pub-lications, encompassing 420 distinct records, this research focuses predominantly on three primary fiber types: crimped, hooked, and mil-cut. Complemented by hands-on experimentation involving 20 diverse SFRC mixtures, this empirical campaign is further illuminated through the strategic use of Partial Dependence Plots (PDPs), revealing intricate relationships between input parameters and con-sequent compressive strength. A pivotal revelation of this research lies in the identification of optimal SFRC formulations, offering tangible insights for real-world applications. The developed ML model stands out not only for its sophistication but also its tangible accuracy, evidenced by exemplary performance against inde-pendent datasets, boasting a commendable mean target-prediction ratio of 99%. To bridge the theory-practice gap, we introduce a user-friendly digital interface, thri-ughly designed to guide professionals in optimizing and accurately predicting the compressive strength of SFRC. This research thus contributes to the construction and civil engineering sectors by enhancing predictive capabilities and refining mix designs, fostering innovation, and addressing the evolving needs of the industry.

In the era of big data, feature engineering has proved its efficiency and importance in dimensionality reduction and useful information extraction from original features. Feature engineering can be expressed as dimensionality reduction and is divided into two types of methods such as feature selection and feature extraction. Each method has its pros and cons. There are a lot of studies to combine these methods. Sparse autoencoder (SAE) is a representative deep feature learning method that combines feature selection with feature extraction. However, existing SAEs do not consider the feature importance during training. It causes extracting irrelevant information. In this paper, we propose a parallel guiding sparse autoencoder (PGSAE) to guide the information by two parallel guiding layers and sparsity constraints. The parallel guiding layers keep the main distribution using Wasserstein distance which is a metric of distribution difference, and it suppresses the leverage of guiding features to prevent overfitting. We perform our experiments using four datasets that have different dimensionality and number of samples. The proposed PGSAE method produces a better classification performance compared to other dimensionality reduction methods.

In the chemical processing industries, sensors for pumps are among the most commonly used machinery. Condition-based maintenance (CBM) and prognosis health management (PHM) determine the most cost-effective time to overhaul pumps. In order to determine the status of the pump, a signal-emitting accelerometer is employed. Stationarity-based feature extraction from amplitude signals is used to process the signal. Utilizing the time-domain function, multiple statistical results were produced. Eight fault codes were classified using support vector machine method. The enormous amount of data points necessitated the use of feature selection. In terms of accuracy, precision, recall, and F1 score, the Chi-square feature selection method exceeds other approaches.

The importance of antibodies in health care and the biotechnology research and development demands not only knowledge of their experimental structures at high resolution, but also practical implementation of this knowledge for both effective and efficient design and production of antibody for its use in both medical and research applications. While the experimental wet-lab approach is usually costly, laborious and time-consuming, computational (dry-lab) approaches, in spite of their intrinsic limitations in comparison with its experimental (wet-lab) counterpart, provide a cheaper and faster alternative option. For the first time, this article reports a comprehensive set of structural electrostatic features extracted from experimentally determined antigen-antibody-related structures, including especially those structural electrostatic features at the interfaces of all experimentally determined antigen-antibody complex structures as of February 29, 2020, to facilitate effective and efficient machine learning-based computational antibody design using currently available experimental structures inside Protein Data Bank.

Cooling fans are one of the critical components of air-forced (AF) dry-type transformers for regulating internal temperatures. Therefore, effective malfunction detection is crucial to maintain the transformer temperature within an acceptable range and prevent overheating. Regular maintenance occurs periodically and issues with cooling fans may arise between maintenance periods, leading to prolonged operation under malfunctioning conditions and potential failures. In addition, utilities typically have online information about whether a fan works or not without providing information about cooling fan malfunctioning circumstances. To address these challenges, this study proposes learning-based online monitoring techniques to detect malfunctions in AF transformer cooling fans. Random forests (RFs) and convolutional neural networks (CNNs) are developed to classify the audio signals from cooling fans into normal and malfunctioning classes. Unlike RFs, which require separate feature extraction, CNNs are trained based on spectrogram images derived from audio signals. Thus, various time-frequency techniques are utilized for feature extraction in RFs. Besides, multiple data augmentation techniques are employed to enhance the dataset size and diversity. Algorithmic performance is optimized through hyperparameter tuning and classifier threshold adjustment. Simulations reveal that CNNs outperform RFs, whereas the latter provides superior interpretability of acoustic features compared to the former.

Star image registration is the most important step in the application of astronomical image differencing, stacking and mosaicking, which requires high robustness, accuracy and real--time of the algorithm, but there is no high--performance registration algorithm in this field. In this paper, we propose a star image registration algorithm that relies only on radial module features (RMF) and rotation angle features (RAF), which has excellent robustness, high accuracy, and good real--time performance. The test results on a large amount of simulated and real data show that the comprehensive performance of the proposed algorithm is significantly better than the four classical baseline algorithms in the presence of rotation, insufficient overlapping area, false stars, position deviation, magnitude deviation and complex sky background, which is a more ideal star image registration algorithm.

The object detection task is usually affected by complex backgrounds. In this paper, a new image object detection method is proposed, which can perform multi-feature selection on multi-scale feature maps. By this method, a bidirectional multi-scale feature fusion network is designed to fuse semantic features and shallow features to improve the detection effect of small objects in complex backgrounds. When the shallow features are transferred to the top layer, a bottom-up path is added to reduce the number of network layers experienced by the feature fusion network, reducing the loss of shallow features. In addition, a multi-feature selection module based on the attention mechanism is used to minimize the interference of useless information on subsequent classification and regression, allowing the network to adaptively focus on appropriate information for classification or regression to improve detection accuracy. Because the traditional five-parameter regression method has severe boundary problems when predicting objects with large aspect ratios, the proposed network treats angle prediction as a classification task. The experimental results on the DOTA dataset, the self-made DOTA-GF dataset and the HRSC 2016 dataset show that, compared with several popular object detection algorithms, the proposed method has certain advantages in detection accuracy.

Measure While Drilling (MWD) refers to a comprehensive array of sensors employed to gather performance data throughout the process of rock drilling in mining blast holes. Multiple MWD metrics are collected by researchers depending on the setup of MWD sensors. Trends in MWD data can be correlated with pre-excavation subsurface conditions. The low-cost, extensive MWD data acquisition resolution of blast holes compared to expensive and sparsely collected exploration drill holes make the former one a desirable method for characterizing pre-excavation subsurface conditions. Nevertheless, the application of MWD technology in open pit mining is constrained by a substantial amount of data that requiring manual analysis within a short time. To address this limitation, computerized data acquisition and interpretation of MWD variables has been used to broadly determine open pit mining lithological boundaries. The results of artificial intelligence algorithms, such as Neural Networks, and Gaussian Processes, to recognize subsurface conditions, such as rock type, from MWD data have consistently returned >90% prediction accuracy. However, methods of evaluating the importance of each MWD feature, such as Principal Component Analysis, has not been optimally used. This review focuses on the progression of intelligent MWD analysis using Machine Learning methods in open-pit mining, to determine the geotechnical, geological, and geochemical subsurface characteristics before open-pit mining excavation.

With the rapid development of the Internet industry, the problem of information overload has arisen due to the abundance of information available online. Recommendation algorithms, as the core of recommendation systems, have been attracting much attention and are a hot topic of research for many experts and scholars. The classical recommendation algorithms are mainly divided into three major categories: collaborative filtering recommendation algorithms, content-based recommendation algorithms, and hybrid-based recommendation algorithms. Although these algorithms are widely used in various fields, with the proliferation of information, these traditional recommendation algorithms are no longer able to meet the needs of the times. To address this issue, recommendation systems have been developed to provide users with personalized and relevant information or products. Despite the wide use of recommendation algorithms, such as collaborative filtering, content-based filtering, and hybrid approaches, traditional recommendation algorithms have limitations and are no longer suitable for meeting the demands of the times. This paper proposes a new recommendation algorithm, SFRRG, that fuses structure and feature information in graph neural networks to improve the performance of the recommendation system in rating prediction. The effectiveness of the proposed algorithm is demonstrated through experiments on various data sets and compared with existing recommendation algorithms.

Recommender systems as an effective information filtering system can be used to obtain information through the user's explicit or implicit behavior. On the one hand finding items that may be of interest to the user. On the other hand, the recommendation facilitates the interaction between the user and the item to increase the revenue. Recommender systems have been widely used in various fields, such as e-commerce, travel recommendation, online books and movies, social networks, etc, which can satisfy the intrinsic implicit needs of users through personalized services. In recent years, the development of deep learning has further improved the performance of recommendation systems. Although these methods improve the performance of the recommendation system, when the number of users and products increases, the recommendation system may face sparsity and cold start problems, and thus cannot achieve personalized recommendations. Knowledge graphs, which are structured data, have become the choice of many algorithms due to the high quality and wide scale of the data, and therefore many recommendation algorithms combined with knowledge graphs have emerged as a popular new direction in recommendation systems. These algorithms are able to preserve the rich connections between different entities. Moreover, when constructing the features of an entity, the entities that are far away from the central entity can also be utilized. Entities are no longer only directly connected to each other. To address the shortcomings of existing recommendation algorithms, this paper designs the recommendation algorithm GPRE using graph neural networks. GPRE focuses on expressing the user's features. The graph neural network provides GPRE with a strong generalization capability for modeling, which can provide long-range semantics between users and entities, as well as selective entity selection in the auxiliary graph neural network. Explicit semantic links are established between remote and central nodes to reduce the introduction of noise. In this paper, experiments are conducted on real-world datasets and the results are compared with baselines. The experimental results show that GPRE performs well on the experimental dataset.

Fundus image registration plays a crucial role in the clinical evaluation of ocular diseases, such as diabetic retinopathy and macular degeneration, necessitating meticulous monitoring. The alignment of multiple fundus images enables the longitudinal analysis of patient progression, widening the visual scope, or augmenting resolution for detailed examinations. Currently, prevalent methodologies rely on feature-based approaches for fundus registration. However, certain methods exhibit high feature point density, posing challenges in matching due to point similarity. This study introduces a novel fundus image registration technique integrating U-Net for the extraction of feature points employing FIVES for its training and evaluation, a novel and large dataset for blood vessels segmentation, prioritizing point distribution over abundance. Subsequently, the method employs medial axis transform and pattern detection to obtain feature points characterized by the Fast Retina Keypoint (FREAK) descriptor, facilitating matching for transformation matrix computation. Assessment of the vessel segmentation achieves 0.7559 for Intersection Over Union (IoU), while evaluation on the Fundus Image Registration Dataset (FIRE) demonstrates the method’s comparative performance against existing methods, yielding a registration error of 0.596 for area under the curve, refining similar earlier methods and suggesting promising performance comparable to prior methodologies.

Myoelectric control is the cornerstone of many assistive technologies used in clinical practice, such as prosthetics and orthoses, and human-computer interaction, such as virtual reality control. Although the performance of such devices exceeds 90\% in controlled environments, myoelectric devices still face challenges in robustness to variability of daily living conditions. Within this survey, the intrisic physiological mechanisms limiting practical implementations of myoelectric devices were explored: the limb position effect and the contraction intensity effect. The degradation of electromyography (EMG) pattern recognition in the presence of these factors was demonstrated on six datasets, where performance was 13% and 20% lower in realistic environments compared to controlled environments for the limb position and contraction intensity effect, respectively. The experimental designs of limb position and contraction intensity literature were surveyed. Current state-of-the-art training strategies and robust algorithms for both effects were compiled and presented. Recommendations for future limb position effect studies include: the collection protocol providing exemplars of 6 positions (four limb positions and three forearm orientations), three-dimensional space experimental designs, transfer learning approaches, and multi-modal sensor configurations. Recommendations for future contraction intensity effect studies include: the collection of dynamic contractions, nonlinear complexity features, and proportional control.

Detailed built-up area information is valuable for mapping complex urban environments. Although a large number of classification algorithms about built-up areas have been developed, they are rarely tested from the perspective of feature engineering and feature learning. Therefore we launched a unique investigation to provide a full test of the OLI imagery for 15-m resolution built-up area classification in 2015, in Beijing, China. Training a classifier requires many sample points, and we propose a method based on the ESA's 38-meter global built-up area data of 2014, Open Street Map and MOD13Q1-NDVI to achieve rapid and automatic generation of a large number of sample points. Our aim is to examine the influence of a single pixel and image patch under traditional feature engineering and modern feature learning strategies. In feature engineering, we consider spectra, shape and texture as the input features, and SVM, random forest (RF) and AdaBoost as the classification algorithms. In feature learning, the convolution neural network (CNN) is used as the classification algorithm. In total, 26 built-up land cover maps were produced. Experimental results show that: (1) the approaches based on feature learning are generally better than those based on feature engineering in terms of classification accuracy, and the performance of ensemble classifiers e.g., RF, is comparable to that of CNN. Two dimensional CNN and the 7 neighborhood RF have the highest classification accuracy of nearly 91%. (2) Overall, the classification effect and accuracy based on image patches are better than those based on single pixels. The features that can highlight the information of the target category (for example, PanTex and EMBI) can help improve classification accuracy.

Containing hundreds of spectral bands (features), hyperspectral images (HSIs) have high ability in discrimination of land cover classes. Traditional HSIs data processing methods consider the same importance for all bands in the original feature space (OFS), while different spectral bands play different roles in identification of samples of different classes. In order to explore the relative importance of each feature, we learn a weighting matrix and obtain the relative weighted feature space (RWFS) as an enriched feature space for HSIs data analysis in this paper. To overcome the difficulty of limited labeled samples which is common case in HSIs data analysis, we extend our method to semisupervised framework. To transfer available knowledge to unlabeled samples, we employ graph based clustering where low rank representation (LRR) is used to define the similarity function for graph. After construction the RWFS, any arbitrary dimension reduction method and classification algorithm can be employed in RWFS. The experimental results on two well-known HSIs data set show that some dimension reduction algorithms have better performance in the new weighted feature space.

Speech emotion recognition remains a heavy lifting in natural language processing. It has strict requirements to the effectiveness of feature extraction and that of acoustic model. With that in mind, a Heterogeneous Parallel Convolution Bi-LSTM model is proposed to address these challenges. It consists of two heterogeneous branches: the left one contains two dense layers and a Bi-LSTM layer, while the right one contains a dense layer, a convolution layer, and a Bi-LSTM layer. It can exploit the spatiotemporal information more effectively, and achieves 84.65%, 79.67%, and 56.50% unweighted average recall on the benchmark databases EMODB, CASIA, and SAVEE, respectively. Compared with the previous research results, the proposed model achieves better performance stably.

Malaria remains an important cause of death, especially in sub-Saharan Africa with about 228 million malaria cases worldwide and an estimated 405,000 deaths in 2019. Currently, malaria is diagnosed in the health facility using a microscope (BS) or rapid malaria diagnostic test (MRDT) and with area where these tools are inadequate the presumptive treatment is performed. Apart from that self-diagnosis and treatment is also practiced in some of the households. With the high-rate self-medication on malaria drugs, this study aimed at computing the most significant features using feature selection methods for best prediction of malaria in Tanzania that can be used in developing a machine learning model for malaria diagnosis. A malaria symptoms and clinical diagnosis dataset were extracted from patients’ files from four (4) identified health facilities in the regions of Kilimanjaro and Morogoro. These regions were selected to represent the high endemic areas (Morogoro) and low endemic areas (Kilimanjaro) in the country. The dataset contained 2556 instances and 36 variables. The random forest classifier a tree based was used to select the most important features for malaria prediction. Regional based features were obtained to facilitate accurate prediction. The feature ranking as indicated that fever is universally the most influential feature for predicting malaria followed by general body malaise, vomiting and headache. However, these features are ranked differently across the regional datasets. Subsequently, six predictive models, using important features selected by feature selection method, were used to evaluate the features performance. The features identified complies with malaria diagnosis and treatment guideline provided with WHO and Tanzania Mainland. The compliance is observed so as to produce a prediction model that will fit in the current health care provision system in Tanzania.

We propose a new ensemble classification algorithm, named Super Random Subspace Ensemble (Super RaSE), to tackle the sparse classification problem. The proposed algorithm is motivated by the Random Subspace Ensemble algorithm (RaSE). The RaSE method was shown to be a flexible framework that can be coupled with any existing base classification. However, the success of RaSE largely depends on the proper choice of the base classifier, which is unfortunately unknown to us. In this work, we show that Super RaSE avoids the need to choose a base classifier by randomly sampling a collection of classifiers together with the subspace. As a result, Super RaSE is more flexible and robust than RaSE. In addition to the vanilla Super RaSE, we also develop the iterative Super RaSE, which adaptively changes the base classifier distribution as well as the subspace distribution. We show the Super RaSE algorithm and its iterative version perform competitively for a wide range of simulated datasets and two real data examples. The new Super RaSE algorithm and its iterative version are implemented in a new version of the R package RaSEn.

Melanoma is widely recognized as one of the most lethal forms of skin cancer, with its incidence showing an upward trend in recent years. Nonetheless, the timely detection of this malignancy substantially enhances the likelihood of patients’ long-term survival. Several computer-based methods have recently been proposed in the pursuit of diagnosing skin lesions at their early stages. Despite achieving some level of success, there still remains a margin of error that the machine learning community considers to be an unresolved research challenge. This study presents a novel framework for the classification of skin lesions. The framework incorporates deep features to generate a highly discriminant feature vector, while also maintaining the integrity of the original feature space. Recent deep models including Darknet53, DenseNet201, InceptionV3, and InceptionResNetV2 are employed in our study for the purpose of feature extraction. Additionally, transfer learning is leveraged to enhance the performance of our approach. In the subsequent phase, the extracted feature information from the chosen pre-existing models is combined, with the aim of preserving maximum information, prior to undergoing the process of feature selection using a novel entropy-controlled grey wolf optimization (ECGWO) algorithm. The integration of fusion and selection techniques is employed to initially incorporate the feature vector with a high level of information and subsequently eliminate redundant and irrelevant feature information. The efficacy of our design is substantiated through the evaluation on three benchmark dermoscopic datasets, namely PH2, ISIC-MSK, and ISIC-UDA. In order to validate the proposed methodology, a comprehensive evaluation is conducted, including a rigorous comparison with established techniques in the field.

The prompt and accurate diagnosis of breast lesions, including the distinction between cancer, non-cancer, and suspicious cancer, plays a crucial role in the prognosis of breast cancer. In this paper, we introduce a novel method based on feature extraction and reduction for detection of breast cancer in mammography images. First, we extract features from multiple pre-trained convolutional neural network (CNN) models, and then concatenate them. The most informative features are selected based on their mutual information with the target variable. Subsequently, the selected features can be classified using a machine learning algorithm. We evaluate our approach using four different machine learning algorithms, and our results demonstrate that the neural network-based classifier yields an accuracy as high as 92% for the RSNA dataset which is a new dataset that provides two views and additional features such as age. We compare our proposed algorithm with state-of-the-art methods and demonstrate its superiority, particularly in terms of accuracy and sensitivity. For the MIAS dataset, we achieve an accuracy as high as 94.5%, and for the DDSM dataset, an accuracy of 96% is attained. These results highlight the effectiveness of our method in accurately diagnosing breast lesions and surpassing existing approaches.

Contemporary advanced models for aspect-targeted opinion word extraction (ATOWE), which predominantly utilize BERT-based encoders at a word level, have shown limited advancements when integrated with graph convolutional networks (GCNs) for syntactic tree assimilation. Recognizing the prowess of BERT subwords in encapsulating rare or context-poor words, this study pivots from syntactic trees to BERT subwords, omitting GCNs from the structural framework. Our approach, named Aspect-Enhanced Wordpiece Extraction Model (AEWEM), focuses on refining aspect representation during encoding. We propose an input format of paired sentence-aspect, diverging from traditional single-sentence inputs. AEWEM demonstrates superior performance on benchmark datasets, establishing a robust foundation for future explorations in this domain.

The internal wave recognition algorithm in an ocean data buoy system can be used to realize the real-time and flexible observation of internal waves, but there is no accurate automatic recognition method. To meet the need for automatic, real-time, and reliable internal wave recognition, an automatic internal wave recognition algorithm has been proposed for a tightly profiled intelligent buoy system. The sea profile temperature data collected by the Bailong buoy system in the Andaman Sea in 2018 were used to train and test the internal wave recognition neural network model, which consists of two parts: feature extraction and feature classification. The experiment compares the long short-term memory network (LSTM), convolutional neural network (CNN) with different layers, and deep neural network (DNN) without a feature extraction network and adjusts the number of convolutional nuclei and convolutional strides to improve the feature extraction efficiency. Experiments show that the best results can be obtained when a CNN layer is used as the feature extraction network, the convolutional step length is 4, the number of convolutional kernels is 5. The recall reaches 95.31% and the precision is 97.53%. The internal wave identification delay of the algorithm is 5.0862 minutes, the number of parameters is 1593, and the number of calculations is 3024. The algorithm can be directly deployed to the ocean data buoy system to realize the demand for automatic, real-time and reliable internal wave identification at the buoy end.

By applying a running average (with a window-size= d), we could transform Discrete data to broad-range, Continuous values. When we have more than 2 columns and one of them is containing data about the tags of classification (Class Column), we could compare and sort the features (Non-class Columns) based on the R2 coefficient of the regression for running averages. The parameters tuning could help us to select the best features (the non-class columns which have the best correlation with the Class Column). “Window size” and “Ordering” could be tuned to achieve the goal. this optimization problem is hard and we need an Algorithm (or Heuristics) for simplifying this tuning. We demonstrate a novel heuristics, Called Simulated Distillation (SimulaD), which could help us to gain a somehow good results with this optimization problem.

Biologists seek to identify a small number of significant features that are important, non-redundant, and relevant from diverse omics data. For example, statistical methods like LIMMA and DEseq distinguish differentially expressed genes between a case and control group from the transcript profile. Researchers also apply various column subset selection algorithms on genomics datasets for a similar purpose. Unfortunately, genes selected by such statistical or machine learning methods are often highly co-regulated, making their performance inconsistent. Here, we introduce a novel feature selection algorithm that selects highly disease-related and non-redundant features from a diverse set of omics datasets. We successfully applied this algorithm to three different biological problems: a) disease to normal sample classification, b) multiclass classification of different disease samples, and c) disease subtypes detection. Considering classification ROC-AUC, False-positive, and False-negative rates, our algorithm outperformed other gene selection and differential expression (DE) methods for all six types of cancer datasets from TCGA considered here for binary and multiclass classification problems. Moreover, genes picked by our algorithm improved the disease subtyping accuracy for four different cancer types over the state-of-the-art methods. Hence, we posit that our proposed feature reduction method can support the community to solve various problems, including the selection of disease-specific biomarkers, precision medicine design, and disease sub-type detection.

Although increasing knowledge about biological systems has advanced exponentially in recent decades, it is surprising to realize that the very definition of Life keeps presenting theoretical challenges. Even if several lines of reasoning seek to identify the essence of life phenomenon, most of these thoughts contain fundamental problem in their basic conceptual structure. Most concepts fail to identify necessary and sufficient features to define life. Here, we analyzed the main conceptual framework regarding theoretical aspects supporting life concepts, such as (i) the physical, (ii) the cellular and (iii) the molecular approaches. Based on ontological analysis, we propose that Life should not be positioned under the ontological category of Matter. Yet, life should be better understood under the top-level ontology of “Process”. Exercising an epistemological approach, we propose that the essential characteristic pervading each and every living being is the presence of organic codes. Therefore, we explore theories in biosemiotics in order to propose a clear concept of life as a macrocode composed by multiple inter-related coding layers. Therefore, we suggest a clear distinction between the concept of life and living beings, a distinction that is not evident in theoretical terms. From the proposed concept, we suggest that the evolutionary process is a fundamental characteristic for life’s maintenance but not to its definition. The current proposition opens a fertile field of debate in astrobiology, biosemiotics and robotics.

In recent years, research has focused on generating mechanisms to assess the levels of subjects' cognitive workload when performing various activities that demand high concentration levels, such as driving a vehicle. These mechanisms have implemented several tools to analyze cognitive workload where the electroencephalographic (EEG) signals are the most used due to its high precision. However, one of the main challenges in the EEG signals implementing is finding the appropriate information to identify cognitive states. Here we show a new feature selection model for pattern recognition using information from EEG signals based on machine learning techniques called GALoRIS. GALoRIS combines Genetic Algorithms and Logistic Regression to create a new fitness function that identifies and selects the critical EEG features that contribute to recognizing high and low cognitive workload and structures a new dataset capable of optimizing the model's predictive process. We found that GALoRIS identifies data related to high and low cognitive workload of subjects while driving a vehicle using information extracted from multiple EEG signals, reducing the original dataset by more than 50%, maximizing the model's predictive capacity-achieving a precision rate greater than 90%.

Feature selection is one of the main data preprocessing steps in machine learning. Its goal is to reduce the number of features by removing extra and noisy features. Feature selection methods must consider the accuracy of classification algorithms while performing feature reduction on a dataset. Meta-heuristic algorithms are the most successful and promising methods for solving this issue. The symbiotic organisms search algorithm is one of the successful meta-heuristic algorithms which is inspired by the interaction of organisms in the nature called Parasitism Commensalism Mutualism. In this paper, three engulfing binary methods based on the symbiotic organisms search algorithm are presented for solving the feature selection problem. In the first and second methods, several S-shaped and V-shaped transfer functions are used for binarizing the symbiotic organisms search algorithm, respectively. These methods are called BSOSS and BSOSV. In the third method, two new operators called BMP and BCP are presented for binarizing the symbiotic organisms search algorithm. This method is called EBSOS. The third approach presents an advanced binary version of the coexistence search algorithm with two new operators, BMP and BCP, to solve the feature selection problem, named EBSOS. The proposed methods are run on 18 standard UCI datasets and compared to base and important meta-heuristic algorithms. The test results show that the EBSOS method has the best performance among the three proposed approaches for binarization of the coexistence search algorithm. Finally, the proposed EBSOS approach was compared to other meta-heuristic methods including the genetic algorithm, binary bat algorithm, binary particle swarm algorithm, binary flower pollination algorithm, binary grey wolf algorithm, binary dragonfly algorithm, and binary chaotic crow search algorithm. The results of different experiments showed that the proposed EBSOS approach has better performance compared to other methods in terms of feature count and accuracy criteria. Furthermore, the proposed EBSOS approach was practically evaluated on spam email detection in particular. The results of this experiment also verified the performance of the proposed EBSOS approach. In addition, the proposed EBSOS approach is particularly combined with the classifiers including SVM, KNN, NB and MLP to evaluate this method performance in the detection of spam emails. The obtained results showed that the proposed EBSOS approach has significantly improved the accuracy and speed of all the classifiers in spam email detection.

This study proposes a new algorithm termed rain cell identification and tracking (RCIT) to identify and track rain cells from high resolution weather radar data. Previous algorithms have limitations when tracking non-consequent rain cells owing to their use of maximum correlation coefficient methods and their lack of an alternative way to handle the variation stages of rain cells during their life cycles. To address these deficiencies, various methods are implemented in the new algorithm. These include the particle image velocimetry (PIV) method for motion estimation and the rain cell matching rule to obtain the stage changes of rain cells. High resolution (5-min and 1-km) radar reflectivity data from three rainy days over the German federal state North Rhine Westphalia (NRW) are used to evaluate the proposed algorithm. The performance of the new algorithm is compared with a radar reflectivity map and verified by two object-oriented methods: structure–amplitude–location (SAL) and geometric index. The verification results suggest that the performance of the new algorithm is good. Application of the RCIT algorithm to the selected cases shows that the inner structure of rainfall events in the experimental region present extreme value distributions, with most rainfall events having a short duration with less intensity. The new algorithm can effectively capture the stage changes of rain cells during their life cycles. The proposed algorithm can serve as the basis for further hydro-meteorological applications such as spatial and temporal analysis of rainfall events and short-term flood forecasting.

Aiming at the problems that are poor generalization performance, low retrieval accuracy and large time consumption of existing content-based image retrieval system, the hierarchical image retrieval method based on multi feature fusion is proposed in this paper. The retrieval accuracy rates on Corel5K, UKbeach and Holidays are 68.23(Top 1), 3.73(N-S) and 88.20(mAp), respectively. The experimental results show that the method proposed in this paper can effectively improve the deficiency of single feature retrieval and save time significantly in the premise of a small amount of loss of accuracy.

Microarray gene expression-based detection and classification of medical conditions have been prominent in research studies over the past few decades. However, extracting relevant data from the high-volume microarray gene expression with inherent nonlinearity and inseparable noise components raises significant challenges during data classification and disease detection. So, this paper proposes a two-level strategy involving feature extraction and selection methods before the classification step. The feature extraction step utilizes Short Term Fourier Transform (STFT), and the feature selection step employs Particle Swarm Optimization (PSO) and Harmonic Search (HS) metaheuristic methods. The classifiers employed are Non-Linear Regression, Gaussian Mixture Model, Softmax Discriminant, Naive Bayes, SVM (Linear), SVM (Polynomial), and SVM (RBF). The two-level extracted relevant features are compared with raw data classification results, including Convolutional Neural Network (CNN) Methodology. Among the methods, STFT with PSO feature selection and SVM (RBF) classifier produced the highest accuracy of 94.47%.

In recent decades, with the ever-growing scale of video data, near-duplicate videos continue to emerge. Data quality issue caused by near-duplicate videos is becoming more and more prominent, which has affected the application of normal videos. Although current studies on near-duplicate video detection can be helpful to uncover data quality issues for videos, they still lack a process of automatic merging for the video data represented by high-dimensional features, which are difficult to automatically clean the near-duplicate videos to improve data quality for video datasets. At present, there are few studies on near-duplicate video data cleaning. The existing studies have the sensitive problems of video data orderliness and clustering initial center under a condition that prior distribution is unknown, which seriously affect the accuracy of near-duplicate video data cleaning. To address the above issues, an automatic near-duplicate video data cleaning method based on a consistent feature hash ring is proposed in this paper. First, a residual network with convolutional block attention modules, a long short-term memory deep network, and an attention model are integrated to construct an RCLA deep network with the multi-headed attention mechanism to extract spatiotemporal features of video data. Then, a consistent feature hash ring is constructed, which can effectively alleviate the sensitivity of video data orderliness while providing a condition of near-duplicate video merging. To reduce the sensitivity of the initial cluster centers to results of near-duplicate video cleansing, an optimized feature distance-means clustering algorithm is constructed by utilizing a mountain peak function on a consistent feature hash ring, which can implement automatic cleaning of near-duplicate video data. Finally, experiments are conducted based on a commonly used dataset named CC_WEB_VIDEO and a coal mining video dataset. Compared with some existing works, simulation results demonstrate that the performance of the proposed method.

The paper demonstrates the analysis of Corona Virus Disease based on a probabilistic model. It involves a technique for classification and prediction by recognizing typical and diagnostically most important CT images features relating to Corona Virus. The main contributions of the research include predicting the probability of recurrences in no recurrence (first time detection) cases at applying our proposed approach for feature extraction. The combination of the conventional statistical and machine learning tools is applied for feature extraction from CT images through four images filters in combination with proposed composite hybrid feature extraction (CHFS). The selected features were classified by the stack hybrid classification system(SHC). Experimental study with real data demonstrates the feasibility and potential of the proposed approach for the said cause.

Gender recognition is trivial for physiotherapist, but it is considered a challenge for computers. The electromyography (EMG) and heart rate variability (HRV) were utilized in this work for gender recognition during the stepping exercise using a stepper. The relevant features were extracted and selected. The selected features were then fused to automatically predict gender recognition. However, the feature selection for gender classification became a challenge to ensure better accuracy. Thus, in this paper, a feature selection approach based on both the performance and the diversity between the two features from the rank-score characteristic (RSC) function in a combinatorial fusion approach (CFA) was employed. Then, the features from the selected feature sets were fused using a CFA. The results were then compared with other fusion techniques such as naive bayes (NB), decision tree (J48), k-nearest neighbor (KNN) and support vector machine (SMO). Besides, the results were also compared with previous researches in gender recognition. The experimental results showed that the CFA was efficient for feature selection. The fusion method was also able to improve the accuracy of the gender recognition rate. The CFA provides much better gender classification results which is 94.51% compared to Nazarloo's work (90.34%) and other classifiers.

Cracks on concrete surfaces are vital factors affecting construction safety. Accurate and efficient crack detection can prevent safety accidents. Using drones to photograph cracks on a concrete surface and detect them through computer vision technology has the advantages of accurate target recognition, simple practical operation, and low cost. To solve this problem, an improved CenterNet concrete crack-detection model is proposed. First, a channel-space attention mechanism is added to the original model to enhance the ability of the convolution neural network to pay attention to the image. Second, a feature selection module is introduced to scale the feature map in the downsampling stage to a uniform size and combine it in the channel dimension. In the upsampling stage, the feature selection module adaptively selects the combined features and fuses them with the output features of the upsampling. Finally, the target size loss is optimised from a Smooth L1 Loss to IoU Loss to improve its inability to adapt to targets of different sizes. The experimental results show that the improved CenterNet model reduces the FPS by 123.7, increases the GPU memory by 62MB, increases the FLOPs by 3.81, and increases the AP by 0.154 compared with the original model. The GPU memory occupancy remained stable during the training process and exhibited good real-time performance and robustness.