Sleep disorder is a disease that can be categorized as both an emotional and physical problem. It imposes several difficulties and problems, such as distress during the day, sleep-wake disorders, anxiety, and several other problems. Hence, the main objective of this research is to utilize the strong capabilities of machine learning in the prediction of sleep disorders. In specific, this research aims to meet three main objectives. These objectives are to identify the best regression model, the best classification model, and the best learning strategy that highly suits sleep disorder datasets. Considering two related datasets and several evaluation metrics that are related to the tasks of regression and classification, the results revealed the superiority of the MultilayerPerceptron, SMOreg, and KStar regression models compared with the other twenty-three regression models. Also, IBK, RandomForest, and RandomizableFilteredClassifier showed superior performance compared with other classification models that belong to several learning strategies. Finally, the Function learning strategy showed the best predictive performance among the six considered strategies in both datasets and with respect to the most evaluation metrics.

Infection of liver flukes (Opisthorchis viverrini) is partly due to their suitability for habitats in sub-basin areas, which causes the intermediate host to remain in the watershed system in all seasons. Spatial monitoring of fluke infection at the small basin analysis scale is important because this can enable analysis at the level of the spatial factors involved and influencing infections. The spatial mathematical model was weighted by the nine spatial factors by dividing the analysis into two levels. 1) sub-basin boundary level analyzed with ordinary least square (OLS) model used to analyze spatial factors of liver fluke aimed at analyzing spatial factors related to human liver fluke infection according to sub-basin boundaries, and 2) infection risk positional analysis level with machine learning-based forest classification and regression (FCR) and displaying predictive results of infection risk locations along stream lines. The analysis results show 4 prototype models that import different independent variable factors. The results show that Model-1 and Model-2 give the most AUC = 0.964 and the variables that influence infection risk the most were distance to stream lines, and distance to water bodies, NDMI and NDVI factors rarely affect accuracy. This FCR machine learning application approach can be applied to the analysis of infection risk areas at the sub-basin level, but independent variables must be screened with a preliminary mathematical model weighted to the spatial units in order to obtain the most accurate predictions.

During a multi-detector computed tomography (MDCT) examination, it is crucial to efficiently organize, store, and transmit medical images in DICOM standard, which requires significant hardware resources and memory. Our project processed large amounts of DICOM images by classifying them based on cross-section views that may carry important information about a possible diagnosis. We ensured that images were retained and saved in PNG format to optimize hardware resources while preserving patient confidentiality. Furthermore, we have developed a graphical, user-friendly interface that allows physicians to visualize specific regions of interest in a patient's brain where changes may indicate disease. Our proposed method enables quick classification of medical images into predefined classes of confirmed diseases of brain parenchyma, contributing to swift decision-making for further diagnosis for more precisely evaluating and characterizing brain changes, and it can lead to the rapid application of adequate therapy, which may result in better outcomes.

Machine learning (ML) is increasingly useful as data grows in volume and accessibility as it can perform tasks (e.g. categorisation, decision making, anomaly detection, etc.) through experience and without explicit instruction, even when the data are too vast, complex, highly variable, full of errors to be analysed in other ways , . Thus, ML is great for natural language, images, or other complex and messy data available in large and growing volumes. Selecting a ML algorithm depends on many factors as algorithms vary in supervision needed, tolerable error levels, and ability to account for order or temporal context, among many other things. Importantly, ML methods for explicitly ordered or time-dependent data struggle with errors or data asymmetry. Most data are at least implicitly ordered, potentially allowing a hidden `arrow of time’ to affect non-temporal ML performance. This research explores the interaction of ML and implicit order by training two ML algorithms on Twitter data before performing automatic classification tasks under conditions that balance volume and complexity of data. Results show that performance was affected, suggesting that researchers should carefully consider time when selecting appropriate ML algorithms, even when time is only implicitly included.

This paper presents a novel method for automatic modulation classification (AMC) for cognitive radio (CR) networks based on a simple classifier that is trained with high-order cumulant. The proposed method focuses on the statistical behavior of modulated signals and includes analog modulation and digital schemes, which received less attention in the literature. The effectiveness of the proposed method is demonstrated through simulation results using high-quality generated signals under different signal-to-noise ratios (SNRs) and channel conditions. The classification performance achieved by the proposed method is superior to that of the more complex deep learning methods, making it well-suited for deployment in end units of CR networks, particularly in military and emergency service applications. The proposed method offers a cost-effective, high-quality solution for AMC that meets the stringent requirements of these critical applications.

Electronic invoicing has become mandatory for Italian companies since January 2019. Invoices are structured in a predefined xml template where the information reported can be easily extracted and analyzed. The main aim of this paper is to exploit the information structured in electronic invoices to build an intelligent system which can facilitate accountants work. More precisely, this contribution shows how it is possible to automate part of the accounting process: all sent or received invoices of a company are classified into specific codes which represent the economic nature of the the financial transactions. In order to classify data contained in the invoices a machine learning multiclass classification problem is proposed using as input variables the information of the invoices to predict two different target variables, account codes and the VAT codes, which composes a general ledger entry. Different approaches are compared in terms of prediction accuracy. The best performance is achieved considering the hierarchical structure of the account codes.

In this paper we use quantum machine learning to detect and classify ocular diseases across age related macular degradation, cataract, diabetic, glaucoma, hypertension, and patological myopia categories versus a control group. We analyze fundus imagery from 1000 patients. Early findings indicate there may be benefit in terms of accuracy and loss function minimization of 2.07% and 1.979x respectively compared to a similar method implemented using traditional computers.

Climbing plants require an external support to grow vertically and enhance light acquisition. Climbers that find a suitable support have greater performance and fitness than those that remain prostrate. Support search is characterized by oscillatory movements (i.e., circumnutation), in which plants rotate around a central axis during their growth. Numerous studies have elucidated the mechanistic details of circumnutation, but how this phenomenon is controlled remains unclear. Here, we use supervised machine learning algorithms to generate models that learn to discriminate between the presence/absence of a support in the environment. Results indicate that there is a difference in the pattern of circumnutation, depending on the presence of a support, that can be learned and classified rather accurately. We also identify distinctive kinematic features that contribute to the classification tasks. Overall, machine learning approaches appear to be powerful tools for understanding the movement of plants.

Regulators require financial institutions to estimate counterparty default risks from liquid CDS quotes for the valuation and risk management of OTC derivatives. However, the vast majority of counterparties do not have liquid CDS quotes and need proxy CDS rates. Existing methods cannot account for counterparty-specific default risks; we propose to construct proxy CDS rates by associating to illiquid counterparty liquid CDS Proxy based on Machine Learning Techniques. After testing 156 classifiers from 8 most popular classifier families, we found that some classifiers achieve highly satisfactory accuracy rates. Furthermore, we have rank-ordered the performances and investigated performance variations amongst and within the 8 classifier families. This paper is, to the best of our knowledge, the first systematic study of CDS Proxy construction by Machine Learning techniques, and the first systematic classifier comparison study based entirely on financial market data. Its findings both confirm and contrast existing classifier performance literature. Given the typically highly correlated nature of financial data, we investigated the impact of correlation on classifier performance. The techniques used in this paper should be of interest for financial institutions seeking a CDS Proxy method, and can serve for proxy construction for other financial variables. Some directions for future research are indicated.

This project focuses on using machine learning classification algorithms to determine whether two people are 6 feet apart or not. Two Raspberry Pis were used simulate smart phones. RSSI values of the Bluetooth beacons transmitted between the Raspberry Pis were collected and recorded to train the classifier. The Gaussian Support Vector Machine Classifer yielded the highest testing accuracy of 79.670 and the Decision Tree Classifier yielded the highest AUC of 0.80.

As Artificial Intelligence (AI) technology, particularly Machine Learning algorithms, becomes increasingly ubiquitous, our abilities to understand and interpret AI and machine learning algorithms are becoming increasingly desirable. Visualization has been used as a common tool to view and understand complex machine learning processes. In this paper, we will focus on developing a general framework for the visualization of machine learning models as a scalar valued multi-dimensional function to help users understand how the models behave over different viewing spaces. We first give a formal definition of the visualization problem. Interpolation-based morphing and subspace sampling techniques are applied to generate various renderings through projections and cross-sections of the model space as 3D surfaces or heatmap images. This method will be applied to two real-world datasets and applications: the diagnosis of Alzheimer's Disease (AD) using a human brain networks dataset and a real-world benchmark dataset for predicting home credit default risks. The visualizations show that different machine learning algorithms can behave quite differently under different conditions.

Evolutionary Machine Learning (EML) has demonstrated its effectiveness in modeling various real-world application domains, including the dynamics of diseases and the creation of predictive models across diverse fields. However, the challenge of data imbalance adversely impacts the performance of EML models. While strategies involving misclassification costs or cost-sensitive learning exist to address this issue, their integration into EML has yet to be thoroughly explored. This paper aims to fill this gap by investigating the application of cost-sensitive EML for intelligent health risk prediction. More specifically, this paper delves into the utilization of cost-sensitive genetic programming (CSGP) and cost-sensitive genetic algorithms (CSGA), which are among the most widely used algorithms in the EML category. The study entails training CSGP and CSGA to predict the early onset of three chronic conditions (hypertension, diabetes, and fatty acid-related disorders) through the development of an EML model. Subsequently, the model's performance is evaluated through standard fitness functions such as sensitivity, specificity, and F1-score. From the experimental results, EML models demonstrate promising results in identifying the occurrence of chronic diseases. Further optimization of the cost-sensitive EML model holds the potential for enhanced outcomes in modeling medical problems.

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.

Medical health systems have been concentrating on artificial intelligence techniques for speedy diagnosis. However, the recording of health data in a standard form still requires attention so that machine learning can be more accurate and reliable by considering multiple parameters. The aim of this study is to develop a general framework for recording diagnostic data in an international standard format to facilitate auto-prediction of eye diseases. Efforts were made to ensure error-free data entry by developing a user-friendly interface. Furthermore, different machine learning algorithms were used to analyze patient data based on multiple parameters, including age, illness history, and clinical observations. This data was formatted according to structured hierarchies designed by medical experts, whereas diagnosis was made as per the ICD-10 coding developed by the American Academy of Ophthalmology. Furthermore, the system is designed to evolve through self-learning by adding new classifications for both diagnosis and symptoms. The classification results from tree-based methods demonstrated that the proposed framework performs satisfactorily given a sufficient amount of data. The random forest and decision tree algorithms predicted more accurately as compared to neural networks and the na{\"i}ve Bayes algorithm owing to structured data arrangement.

In this paper we describe a method which combines sentiment analysis with machine learning techniques and/or multivariate statistical analysis. By applying this methodology it is possible to classify a collection of texts into two or more groups or clusters. On the basis of a number of previously defined clusters, the novelty of the outlined approach is the use of the sentiment analysis results as input to the machine learning model or multivariate statistical analysis. Once the classifier has been obtained, we can assign a given text into one of the pre-established clusters. The groups or clusters can represent different time periods, classes of texts transcribed from different conversations, etc. The method is illustrated through an example taken from one of the two studies in which we have applied this methodology. In one of the studies, the method was used to classify press news of a volcanic eruption, while in the other study it was used to classify the conversations recorded between a chatbot with different kinds of speakers (humans or chatbots). This last study was the seminal work in which we introduced this methodology.

The exponential growth of digital media content has introduced new challenges in managing and classifying internet traffic. Digital media traffic is composed of various applications such as video, audio, social media, and search, and its data structure is complex, incorporating a vast array of features. The classification of traffic data is a crucial aspect of internet traffic management and network security, and it forms the basis for several scenarios, including content distribution, advertising recommendations, and data analysis. Traditional classification methods rely mainly on deep packet inspection and port-based techniques, which have become increasingly ineffective due to the rapid evolution of network traffic. To address this issue, this study proposes a machine learning-based traffic classification method aimed at enhancing the accuracy and efficiency of digital media traffic classification to meet the current needs of traffic management and network security. The paper also analyzes and evaluates the classification effect and prediction capability of various algorithms under different training set sizes to validate the feasibility and effectiveness of the proposed method. The result demonstrates that the neural network algorithm has superior classification and prediction capabilities compared to the decision tree and support vector machine algorithms. Furthermore, our proposed method achieves the highest accuracy of 96.88% with a large training sample of 40,000 data streams, proving its superiority in handling high-dimensional data and complex datasets. The research results are significant for the development of digital media traffic classification and prediction methods and are expected to be applied in practical scenarios.

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%.

Machine learning (ML) is a subdivision of artificial intelligence in which the machine learns from machine-readable data and information. It uses data, learns the pattern and predicts the new outcomes. Its popularity is growing because it helps to understand the trend and provides a solution that can be either a model or a product. Applications of ML algorithms have increased drastically in G.I.S. and remote sensing in recent years. It has a broad range of applications, from developing energy-based models to assessing soil liquefaction to creating a relation between air quality and mortality. Here, in this paper, we discuss the most popular supervised ML models (classification and regression) in G.I.S. and remote sensing. The motivation for writing this paper is that ML models produce higher accuracy than traditional parametric classifiers, especially for complex data with many predictor variables. This paper provides a general overview of some popular supervised non-parametric ML models that can be used in most of the G.I.S. and remote sensing-based projects. We discuss classification (Naïve Bayes (NB), Support Vector Machine (SVM), Random Forest (RF), Decision Trees (DT)) and regression models (Random Forest (RF), Support Vector Machine (SVM), Linear and Non-Linear) here. Therefore, the article can be a guide to those interested in using ML models in their G.I.S. and remote sensing-based projects

Soccer is type of sport that carries a high risk of injury. Injury is not only cause in the unlucky soccer carrier and also team performance as well as financial effects can be worse since soccer is a team-based game. The duration of recovery from a soccer injury typically relies on its type and severity. Therefore, we conduct this research in order to predict the probability of players injury type using machine learning technologies in this paper. Furthermore, we compare different machine learning models to find the best fit model. Supervised classification machine learning models are applied in this paper. We gathered information about 54 professional soccer players who are playing in the top five European leagues based on their career history.

Hepatitis C presents a significant global health challenge, necessitating early diagnosis and precise severity classification for timely medical intervention. This study explores the application of machine learning techniques to predict the severity of hepatitis C, leveraging an extensive dataset. Our approach encompasses rigorous data preprocessing, advanced model development, and fine-tuned hyperparameter optimization to ensure accurate and reliable predictions. We evaluated four classification models: Logistic Regression, Random Forest, Gradient Boosting, and Support Vector Machine (SVM), comparing their accuracy in classifying patients. The results showed that the Random Forest and Gradient Boosting models outperformed the others with an accuracy of approximately 93.50%, demonstrating their potential in assisting Hepatitis C diagnosis. Further model enhancements through hyperparameter tuning and feature engineering can improve the precision of Hepatitis C diagnosis, contributing to better patient care.

Water quality assessments are crucial for human health and environmental safeguards. The utilization of a subset of artificial intelligence such as Machine Learning (ML) presents significant impacts to enhance the prediction and classification of water quality. In this research, a set of diverse ML algorithms was evaluated to handle a comprehensive dataset of water quality measurements over an extended period. The aim was to develop a robust approach for accurately forecasting water quality. This approach employed machine learning classifiers such as Logistic Regression (LR), Support Vector Machine (SVM), Stochastic Gradient Descent (SGD), K-Nearest Neighbors (KNN), Gaussian Process Classification (GPC), Gaussian Naive Bayes (GNB), Random Forest (RF), Decision Tree (DT), XGBoost, and Multilayer Perceptron (MLP). The water quality parameters assessed for pH, hardness, solids, chloramines, sulfate, conductivity, organic carbon, trihalomethanes and turbidity. The XGBoost model exhibited the highest accuracy of 89.47% among the classifiers and Stacked Ensemble Classifiers (SEC) improved the prediction further to 92.98%. The findings suggest that XGBoost and the SEC hold promise as reliable approaches for water quality assessments in contrast of artificial intelligence.

In this article, we investigate the applicability of quantum machine learning for classification tasks using two quantum classifiers from the Qiskit Python environment: the Variational Quantum Classifier (VQC) and the Quantum Kernel Estimator (QKE). We evaluate the performance of these classifiers on six widely known and publicly available benchmark datasets and analyze how their performance varies with the number of samples on two artificially generated test classification datasets. As quantum machine learning is based on unitary transformations, this paper explores data structures and application fields that could be particularly suitable for quantum advantages. Hereby, we developed a data set based on concepts from quantum mechanics using the exponential map of a Lie algebra. This dataset will be made publicly available and contributes a novel contribution to the empirical evaluation of quantum supremacy. We further compared the performance of VQC and QKE on six widely applicable datasets to contextualize our results.\\ Our results demonstrate that the VQC and QKE perform better than basic machine learning algorithms such as advanced linear regression models (Ridge and Lasso). They do not match the accuracy and runtime performance of sophisticated modern boosting classifiers like XGBoost, LightGBM, or CatBoost. Therefore, we conclude that while quantum machine learning algorithms have the potential to surpass classical machine learning methods in the future, especially when physical quantum infrastructure becomes widely available, they currently lag behind classical approaches. Our investigations also show that classical machine learning approaches have superior performance classifying datasets based on group structures, compared to quantum approaches that particularly use unitary processes.\\ Furthermore, our findings highlight the significant impact of different quantum simulators, feature maps, and quantum circuits on the performance of the employed quantum estimators. This observation emphasizes the need for researchers to provide detailed explanations of their hyperparameter choices for quantum machine learning algorithms, as this aspect is currently overlooked in many studies within the field.\\ To facilitate further research in this area and ensure the transparency of our study, we have made the complete code available in a linked GitHub repository.

Olive Anthracnose (OA) is the most important fungal disease of olive fruits worldwide. In the context of integrated pest management, the development of predictive models could be used for early diagnosis and control. In the current study, a dataset representing 58 cases (6 locations with 12 olive cultivars) was used to study the relationship between ΟΑ incidence (OAI) and 35 heterogeneous variables, including orchard characteristics, olive fruit parameters, foliar and soil nutrients, soil parameters and soil texture classes. The Random Forest-Recursive Feature Elimination with Cross Validation (RF-RFECV) feature selection method identified Location, Water Content, P, Ca, Mg, Exchangeable Mg, Trace Zn, Trace Cu as possible new indicators associated with OAI. Six different classification algorithms, namely Decision Tree (DT), Gradient Boosting (GB), Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbors (KNN) and Support Vector Machine (SVM), were developed for predicting conditions leading to OAI >0% and 10%. Hyperparameter optimization using grid search was used to optimize the parameters of the models and finally the best parameters were applied to predict the OAI. The final models were evaluated in terms of several standard metrics, such as accuracy, sensitivity, specificity and ROC AUC score. Findings suggested that GB performance was superior compared to the other models for the prediction of the occurrence of OA disease (OAI>0%) with an accuracy of 86.7%, a sensitivity of 100%, a specificity of 75% and a ROC-AUC score of 93%, while for the prediction of the spread of the disease (OAI>10%), DT stood out with an accuracy of 86.7%, a sensitivity of 81.8%, a specificity of 100% and a ROC-AUC score of 91%. RF classifier performed very well in both cases, with an accuracy of 80%, a sensitivity of 85.7%, a specificity of 75% and a ROC-AUC score of 93% for the prediction of the occurrence of the disease (OAI>0%), and an accuracy of 86.7%, a sensitivity of 90.9%, a specificity of 75% and a ROC-AUC score of 84% for the prediction of the spread of the disease (OAI>10%).

In response to the growing challenges in drone security and airspace management, this study introduces an advanced drone classifier, capable of detecting and categorizing Unmanned Aerial Vehicles (UAVs) based on acoustic signatures. Utilizing a comprehensive database of drone sounds across EU-defined classes (C0 to C3), this research leverages machine learning (ML) techniques for effective UAV identification. The study primarily focuses on the impact of data augmentation methods—pitch shifting, time delays, harmonic distortion, and ambient noise integration—on classifier performance. These techniques aim to mimic real-world acoustic variations, thus enhancing the classifier's robustness and practical applicability. Results indicate that moderate levels of augmentation significantly improve classification accuracy. However, excessive application of these methods can negatively affect performance. The study concludes that sophisticated acoustic data augmentation can substantially enhance ML-driven drone detection, providing a versatile and efficient tool for managing drone-related security risks. This research contributes to UAV detection technology, presenting a model that not only identifies but also categorizes drones, underscoring its potential for diverse operational environments.

Wet granulation is a frequent process in the pharmaceutical industry. As a starting point for numerous dosage forms, the quality of the granulation not only affects subsequent production steps but also impacts the quality of the final product. It is thus crucial and economical to monitor this operation thoroughly. Here, we report on identifying different phases of a granulation process using a machine learning approach. The phases reflect the water content which in turn influences the processability and quality of the granule mass. We used two kinds of microphones and an acceleration sensor to capture acoustic emissions and vibrations. We trained convolutional neural networks (CNNs) to classify the different phases using transformed sound recordings as the input. We achieved a classification accuracy of up to 90 % using vibrational data and an accuracy of up to 97 % using the audible microphone data. Our results indicate the suitability of using audible sound and machine learning to monitor pharmaceutical processes. Moreover, since recording acoustic emissions is contactless, it readily complies with legal regulations and present Good Manufacturing Practices.

Efficient triaging and referral assessment are critical in ensuring prompt medical intervention in the Community Health Care (CHC) system. However, the existing triaging systems in many Community Health Services are an intensive, time-consuming process and often lack accuracy, particularly for various symptoms that might represent heart failure or other health-threatening conditions. There is a noticeable limit of research papers describing AI technologies for triaging patients. This paper proposes a novel quantitative data-driven approach using machine learning (ML) modelling to improve the community clinical triaging process. Furthermore, this study aims to employ the feature selection process and machine learning power to reduce the triaging process’s waiting time and increase accuracy in clinical decision-making. The model was trained on medical records of “Heart Failure” patients’ dataset, which included demographics, past medical history, vital signs, medications, and clinical symptoms. A comparative study was conducted using the list of machine learning algorithms where XGBoost demonstrated the best performance among other ML models. The triage levels of 2,35,982 patients achieved an accuracy of 99.94%, precision of 0.9986, recall of 0.9958, and f1-score of 0.9972 within 0.059 seconds. The proposed diagnostic model can be implemented for the CHC decision system and be developed further for other medical conditions.

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.

It is difficult to obtain necessary information accurately from Social Networking Service (SNS) while raising children, and it is thought that there is a certain demand for the development of a system that presents appropriate information to users according to the child's developmental stage. There are still few examples of research on knowledge extraction that focuses on childcare. This research aims to develop a system that extracts and presents useful knowledge for people who are actually raising children, using texts about childcare posted on Twitter. In many systems numbers in text data are just strings like words and are normalized to zero or simply ignored. In this paper, we created a set of tweet texts and a set of profiles created according to the developmental stages of infants from "0-year-old child" to "6-year-old child". For each set, we used ML algorithms such as NB (Naive Bayes), LR (Logistic Regression), ANN (Approximate Nearest Neighbor algorithms search), XGboost, RF (random forest), decision trees, and SVM (Support Vector Machine) to compare with BERT, a neural language model, to construct a classification model that predicts numbers from "0" to "6" from sentences. The accuracy rate predicted by the BERT (Bidirectional Encoder Representations from Transformers) classifier was slightly higher than that of the NB, LR, ANN, XGboost, RF, decision trees, and SVM classifiers, indicating that the BERT classification method was better.

It is difficult to obtain necessary information accurately from Social Networking Service (SNS) while raising children, and it is thought that there is a certain demand for the development of a system that presents appropriate information to users according to the child's developmental stage. There are still few examples of research on knowledge extraction that focuses on childcare. This research aims to develop a system that extracts and presents useful knowledge for people who are actually raising children, using texts about childcare posted on Twitter. In many systems numbers in text data are just strings like words and are normalized to zero or simply ignored. In this paper, we created a set of tweet texts and a set of profiles created according to the developmental stages of infants from "0-year-old child" to "6-year-old child". For each set, we used Support Vector Machine (SVM), and Bidirectional Encoder Representations from Transformers (BERT), a neural language model, to construct a classification model that predicts numbers from "0" to "6" from sentences. The accuracy rate predicted by the BERT classifier was slightly higher than that of the SVM classifier, indicating that the BERT classification method was better.

Psychomotor developmental delay in infants includes failure to acquire abilities such as sitting, walking, grasping objects and communication at the ages when most infants have acquired these abilities. Known risk factors include a large number of aspects of family environment, socioeconomic position, problems in pregnancy and birth, and maternal health. It is clinically useful to be able to screen for developmental delay so that healthcare interventions can be considered. The present research used machine learning (random forest) to create an algorithm predicting psychomotor delay in 9-month-old infants using information ascertainable at birth and in early infancy. The dataset was the UK longitudinal Millennium Cohort study. Fifty-two predictors measuring socioeconomic indicators, paternal, family and social support for the mother, beliefs about good parenting, maternal health, pregnancy and birth were included in the initial algorithm. Feature reduction showed that of the 52 variables, birthweight, family income and parents’ ages had the highest feature importance scores and could alone correctly predict developmental delay with over 99% sensitivity and 100% specificity. The relationships between delay and some of the predictors, particularly income, were nonlinear and complex. The results suggest that the risk of psychomotor developmental delay can be identified in early infancy using machine learning, and that the best predictors are factors present prior to birth. Surprisingly, the most important factors included in the present study did not include illnesses during pregnancy such as eclampsia and infections.

Automation is at the core of modern industry. It aims to increase production rates, decrease production costs, and reduce human intervention in order to avoid human mistakes and time delays during manufacturing. On the other hand, human assistance is usually required to customize products and reconfigure control systems through a special process interface called Human Machine Interface (HMI). Machine Learning (ML) algorithms can effectively be used to resolve this tradeoff between full automation and human assistance. This paper provides an example of the industrial application of ML algorithms to help human operators save their mental effort and avoid time delays and unintended mistakes for the sake of high production rates. Based on real-time sensor measurements, several ML algorithms have been tried to classify paper rolls according to paper grammage in a white paper mill. The performance evaluation shows that the AdaBoost algorithm is the best ML algorithm for this application with classification accuracy (CA), precision, and recall of 97.1%. The generalization of the proposed approach for achieving a cost-effective mill construction by reducing the total number of the required physical sensors will be the subject of our future research.

Salt marshes are globally important ecosystems, but many have been lost or transformed due to the impacts of global change. There have been attempts to broadly quantify salt marsh communities, especially the ubiquitous grasses which serve as foundation species such as Spartina alterniflora and Spartina patens, the latter of which is being lost due to sea level rise. However, few researchers have used high-resolution geospatial imagery to quantify fine-scale changes in the distribution of grasses or to track losses of S. patens. To address this issue, we utilized a simple and rapid method of classifying geospatial marsh imagery with cloud-based machine learning in Google Earth Engine (>92% accuracy for S. patens regardless of imagery age). Our methods allowed us to characterize full landscapes (two geospatially proximal areas, >7,000 ha each) of critical salt marshes on the New Jersey coast and to evaluate fine-scale (1-m) community transformations in response to global change with imagery from 2006 to 2019. Notably, one marsh experienced very little change while the other experienced an 81.17% (1,087 ha) loss of S. patens, illuminating disparate patterns of change for two geographically proximal ecosystems. Further exploration revealed an association in the loss of S. patens with increases in streamflow and total nitrogen content in the rivers that run through each marsh. These results signify the importance of broad-scale ecological studies that evaluate fine-scale community transformations and for management strategies that do not generalize across landscapes of an ecosystem-type.

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.

Successful engineering of a microbial host for efficient production of a target product from a given substrate can be viewed as an extensive optimization task. Such a task involves the selection of high activity enzymes as well as their gene expression regulatory control elements (i.e., promoters and ribosome binding sites). Finally, there is also the need to tune expression of multiple genes along a heterologous pathway to relieve constraints from rate-limiting step and help reduce metabolic burden on cells from unnecessary over-expression of high activity enzymes. While the aforementioned tasks could be performed through combinatorial experiments, such an approach incurs significant cost, time and effort, which is a handicap that can be relieved by application of modern machine learning tools. Such tools could attempt to predict high activity enzymes from sequence, but they are currently most usefully applied in classifying strong promoters from weaker ones as well as combinatorial tuning of expression of multiple genes. This perspective reviews the application of machine learning tools to aid metabolic pathway optimization through identifying challenges in metabolic engineering that could be overcome with the help of machine learning tools.

Atrial Fibrillation (AF) is known as one of the dominant cardiac disease that is associated with severe consequences such as risk of strokes. This cardiac arrhythmia can be efficiently diagnosed using EEG signals. However, AF episodes in EEG records are not always permanent for all subjects. In addition, lengthy subject preparation and ECG recording sessions makes this practice time consuming and not attractive for efficient diagnosis. One attractive solution for this problem will be the use of wearable Internet of Medical Things IoMT devices for EEG recording and AF detection. From another side the use machine learning for efficient diagnosis could provide high accurate diagnosis. However, the deployment of these wireless wearable devices faces the problems of limited processing capabilities and of energy scarcity. In this context, the design of energy efficient accurate on-sensor AF detection scheme using machine learning algorithms still requires a focused research effort. This paper proposes a wearable solution for ECG-based processing scheme for embedded AF detection. The proposed method achieved 100% sensitivity and 94.5% of specificity and accuracy of around 97.4 % for wide range of subjects while showing low energy-consumption. The obtained results show the efficiency of using Logistic Regression algorithm for accurate AF detection and attest about practical feasibility of the proposed approach in wearable IoMT devices.

The spread of the Coronavirus pandemic has been accompanied by an infodemic. The false information that is embedded in the infodemic affects people’s ability to have access to safety information and follow proper procedures to mitigate the risks. This research aims to target the falsehood part of the infodemic, which prominently proliferates in news articles and false medical publications. Here, we present NeoNet, a novel supervised machine learning text mining algorithm that analyzes the content of a document (news article, a medical publication) and assigns a label to it. The algorithm is trained by TFIDF bigram features which contribute a network training model. The algorithm is tested on two different real-world datasets from the CBC news network and Covid-19 publications. In five different fold comparisons, the algorithm predicted a label of an article with a precision of 97-99 %. When compared with prominent algorithms such as Neural Networks, SVM, and Random Forests NeoNet surpassed them. The analysis highlighted the promise of NeoNet in detecting disputed online contents which may contribute negatively to the COVID-19 pandemic.

In this paper, we examine the performance of several classifiers in the process of searching for very negative opinions. More precisely, we do an empirical study that analyzes the influence of three types of linguistic features (n-grams, word embeddings, and polarity lexicons) and their combinations when they are used to feed different supervised machine learning classifiers: Support Vector Machine (SVM), Naive Bayes (NB), and Decision Tree (DT).

Ubiquitous computing comprises scenarios where networks, devices within the network, and software components change frequently. Market demand and cost-effectiveness are forcing device manufacturers to introduce new-age devices. Also, the Internet of Things (IoT) is transitioning rapidly from the IoT to the Internet of Everything (IoE). Due to this enormous scale, effective management of these devices becomes vital to support trustworthy and high-quality applications. One of the key challenges of IoT device management is automatic device classification with the logically semantic type and using that as a parameter for device context management. This would enable smart security solutions. In this paper, a device classification approach is proposed for the context management of ubiquitous devices based on unsupervised machine learning. To classify unknown devices and to label them logically, a proactive device classification model is framed using a k-Means clustering algorithm. To group devices, it uses the information of network parameters such as Received Signal Strength Indicator (rssi), packet_size, number_of_nodes in the network, throughput, etc. Experimental analysis suggests that the well-formedness of clusters can be used to derive cluster labels as a logically semantic device type which would be a context for resource management and authorization of resources. This paper fulfills an identified need of proactive device classification for device management.

In Southern Asia, Bangladesh is a well-known developing country. Because of COVID-19, we continuously face challenges. Not only can these issues occur beyond economic or health concerns, but they also generate dangerous social problems, such as family abuse. Since the inception of this epidemic, multiple social crimes are looming. Remaining home during the lockout period enhances divorce rates. This research presents a customized forecast of family violence during the COVID-19 outbreak by using machine learning methods. In this paper, we have applied Random Forest, Logistic Regression, and Naive Bayes machine learning classifiers to predict family violence and discovered the feature importance. The performance of the classifiers is evaluated based on accuracy, precision, recall, and F-score. We have employed an oversampling strategy named synthetic minority oversampling technique (SMOTE) to solve the imbalance problem of our data. Even, we have tried to compare three machine learning model performances before and after balancing of normalization data. Finally, ROC analyses and confusion matrices were developed and analyzed by using data augmentation. Our proposed system with the random forest classifier performed better with 77% accuracy in comparison with other two machine learning classifiers.

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.

1) Background: One of the most common cancer that affects men worldwide and North American men is prostate cancer. Gleason score is a pathological grading system to examine the potential aggressiveness of the disease in the prostate tissue. The advancement in computing and next-generation sequencing technology now allow us to study the genomic profiles of patients in association with their different Gleason score more accurately and effectively. 2) Methods: In this study, we used a novel machine learning method to analyze gene expression of prostate tumors with different Gleason scores, and identify potential genetic biomarkers for each Gleason group. We obtained a publicly-available RNA-Seq dataset of a cohort of 104 prostate cancer patients from the National Center for Biotechnology Information’s (NCBI) Gene Expression Omnibus (GEO) repository, and categorized patients based on their Gleason scores to create a hierarchy of disease progression. A hierarchical model with standard classifiers in different Gleason groups, also known as nodes, was developed to identify and predict nodes based on their mRNA or gene expression. In each node, patient samples were analyzed via class imbalance and hybrid feature selection techniques to build the prediction model. The outcome from analysis of each node is a set of genes that can differentiate each Gleason group from the remaining groups. To validate the proposed method, the set of identified genes are used to classify a second dataset of 499 prostate cancer patients collected from cBioportal [1]. 3) Results: The overall accuracy of applying this novel method to the first dataset was 93.3%, and further validated to 87% accuracy using the second dataset. This method also identified genes that were not previously reported as potential biomarkers for specific Gleason groups. In particular, PIAS3 was identified as a potential biomarker for Gleason score 4+3=7, and UBE2V2 for Gleason score 6. 4) Insight: Previous reports show that the genes predicted by this newly proposed method strongly correlate with prostate cancer development and progression. Furthermore, pathway analysis shows that both PIAS3 and UBE2V2 share similar protein interaction pathways, the JAK/STAT signaling process.

This study provides a comprehensive assessment of Derawan Island's coral reefs over two decades (2003, 2011, and 2021) from a machine learning classification perspective. Employing non-parametric algorithms like Random Forest (RF), Support Vector Machine (SVM), and Classification and Regression Tree (CART), our analysis primarily focused on spatial and temporal changes in coral habitats. RF emerged as the most accurate method, demonstrating an accuracy of 71.43% with Landsat 9, 73.68% with Sentinel 2, and 78.28% with Multispectral Aerial Photos. We found that the accuracy of our classification results was significantly influenced by the geographic resolution, as well as the quality of field and satellite/aerial image data. Through spatial clustering, the coral habitats exhibited an Nearest Neighbor Index (NNI) value of 0.8727, indicating specific patterns of distribution. The analysis revealed a decrease in coral reef extent from 2003 to 2011, shrinking to 16 hectares with varying densities, followed by a slight area increase but with more heterogeneous densities between 2011 and 2021. This study not only highlights the dynamic nature of coral reef habitats over two decades but also underscores the critical role of machine learning in environmental monitoring and conservation efforts.

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.

One of the most important challenges in the Machine and Deep Learning areas today is to build good models using small datasets, because sometimes it is not possible to have large ones. Several techniques have been proposed in the literature to address this challenge. This paper aims at studying the different available Deep Learning techniques and performing a thorough experimentation to analyze which technique or combination thereof improves the performance and effectiveness of the models. A complete comparison with classical Machine Learning techniques was carried out, to contrast the results obtained using both techniques when working with small datasets. Thirteen algorithms were implemented and trained using three different small datasets (MNIST, Fashion MNIST, and CIFAR-10). Each experiment was evaluated using a well-established set of metrics (Accuracy, Precision, Recall, F1, and the Matthews correlation coefficient). The experimentation allowed concluding that it is possible to find a technique or combination of them to mitigate a lack of data, but this depends on the nature of the dataset, the amount of data, and the metrics used to evaluate them.

. 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.

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.

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.

Background: the primary care service in Catalonia has operated an asynchronous teleconsulting service between GPs and patients since 2015 (eConsulta), which has generated some 500,000 messages. New developments in big data analysis tools, particularly those involving natural language, can be used to accurately and systematically evaluate the impact of the service. Objective: the study was intended to examine the predictive potential of eConsulta messages through different combinations of vector representation of text and machine learning algorithms and to evaluate their performance. Methodology: 20 machine learning algorithms (based on 5 types of algorithms and 4 text representation techniques)were trained using a sample of 3,559 messages (169,102 words) corresponding to 2,268 teleconsultations (1.57 messages per teleconsultation) in order to predict the three variables of interest (avoiding the need for a face-to-face visit, increased demand and type of use of the teleconsultation). The performance of the various combinations was measured in terms of precision, sensitivity, F-value and the ROC curve. Results: the best-trained algorithms are generally effective, proving themselves to be more robust when approximating the two binary variables "avoiding the need of a face-to-face visit" and "increased demand" (precision = 0.98 and 0.97, respectively) rather than the variable "type of query"(precision = 0.48). Conclusion: to the best of our knowledge, this study is the first to investigate a machine learning strategy for text classification using primary care teleconsultation datasets. The study illustrates the possible capacities of text analysis using artificial intelligence. The development of a robust text classification tool could be feasible by validating it with more data, making it potentially more useful for decision support for health professionals.

In cricket, particularly the twenty20 format is most watched and loved by the people, where no one can guess who will win the match until the last ball of the last over. In India, The Indian Premier League (IPL) started in 2008 and now it is the most popular T20 league in the world. So we decided to develop a machine learning model for predicting the outcome of its matches. Winning in a Cricket Match depends on many key factors like a home ground advantage, past performances on that ground, records at the same venue, the overall experience of the players, record with a particular opposition, and the overall current form of the team and also the individual player. This paper briefs about the key factors that affect the result of the cricket match and the regression model that best fits this data and gives the best predictions. Cricket, the mainstream and widely played sport across India which has the most noteworthy fan base. Indian Premier League follows 20-20 format which is very unpredictable. IPL match predictor is a ML based prediction approach where the data sets and previous stats are trained in all dimensions covering all important factors such as: Toss, Home Ground, Captains, Favorite Players, Opposition Battle, Previous Stats etc, with each factor having different strength with the help of KNIME Tool and with the added intelligence of Naive Bayes network and Eulers strength calculation formula.

This research assesses the classification of Arabic coffee into three primary variations (light, medium, and dark) using simulated data based on actual measurements of color information, antioxidant laboratory tests, and chemical composition tests. Two types of simulated data were generated, with the standard deviation of the measures varied. Multiple classifiers were used for classification. The findings demonstrate the use of color information in accurately distinguishing Arabic coffee types. The different CIE color values, as well as the excellent connection between color information and coffee classes, lead to flawless classification results. However, depending primarily on antioxidant information results in lower classification performance, with variability amongst classifiers due to increasing data complexity. Chemical composition information, on the other hand, possesses outstanding discriminatory strength, allowing faultless classification on its own. Certain characteristics, such as crude protein and crude fiber, show strong correlations and are critical in the classification of coffee types. Based on the findings, it is recommended to create a mobile application that uses image recognition to analyze coffee color while providing chemical composition information. End users, such as consumers, will be able to make informed judgments regarding their coffee preferences owing to this application.

Landslides, resulting from disturbances in slope equilibrium, pose a significant threat to landscapes, infrastructure, and human life. Triggered by factors such as intense precipitation, seismic activities, or volcanic eruptions, these events can cause extensive damage and endanger nearby communities. A comprehensive understanding of landslide characteristics, including spatio-temporal patterns, dimensions, and morphology, is vital for effective landslide-disaster management. Existing Remote Sensing approaches mostly use either optical or Synthetic Aperture Radar (SAR) sensors. Integrating information from both these types of sensors promises greater accuracy for identifying and locating landslides. This study proposes a novel approach, the ML-LaDeCORsat (Machine Learning-based coseismic Landslide Detection using Combined Optical and Radar Satellite Imagery), that integrates freely available Sentinel-1, Palsar-2, and Sentinel-2 imagery data, along with relevant spectral indices and suitable bands using machine-learning-based classification for coseismic landslide detection implemented in Google Earth Engine (GEE). The approach includes a robust and reproducible training and validation strategy. Using landslides from four different earthquake case studies, we demonstrate the superiority of our approach over existing solutions in coseismic landslide iden-tification and localization, providing a detection accuracy of 87-92%. ML-LaDeCORsat can be adapted to other landslide events (GEE script is provided). Transfer learning experiments proofed that our model can be applied to other coseismic landslide events without the need for additional training data. Our novel approach therefore facilitates quick and reliable identification of co-seismic landslides, highlighting its potential to contribute towards more effective disaster management.

Emotion detection from the text is an important and challenging problem in text analytics. The opinion-mining experts are focusing on the development of emotion detection applications as they have received considerable attention of online community including users and business organization for collecting and interpreting public emotions. However, most of the existing works on emotion detection used less efficient machine learning classifiers with limited datasets, resulting in performance degradation. To overcome this issue, this work aims at the evaluation of the performance of different machine learning classifiers on a benchmark emotion dataset. The experimental results show the performance of different machine learning classifiers in terms of different evaluation metrics like precision, recall ad f-measure. Finally, a classifier with the best performance is recommended for the emotion classification.

Current conventional methods of evaluating microstructures are characterized by a high degree of subjectivity and a lack of reproducibility. Modern machine learning (ML) approaches have already shown great potential in overcoming these challenges. Once trained with representative data in combination with an objective ground truth, the ML model is able to perform a task properly in a reproducible and automated manner. However, in highly complex use cases, it is often not possible to create a definite ground truth. This study addresses this problem using the underlying showcase of microstructures of highly complex quenched and quenched and tempered (Q/QT) steels. A patch-wise classification approach combined with a sliding window technique provides a solution for segmenting entire microphotographs where pixel-wise segmentation is not applicable due to the problem of reproducibly creating unambiguous training masks. Using correlative microscopy, consisting of light optical microscope (LOM) and scanning electron microscope (SEM) micrographs as well as corresponding data from electron backscatter diffraction (EBSD), a training dataset of reference states that covers a wide range of microstructures was acquired in order to train accurate and robust ML models. Despite the enormous complexity associated with the steels treated here, classification accuracies of 88.8% in the case of LOM images and 93.7% for high-resolution SEM images were achieved. These high accuracies are close to super-human performance, especially in consideration of the reproducibility of the automated ML approaches compared to conventional methods based on subjective evaluations through experts.

The paper aims to develop and test an algorithm for classifying Russian-language text reviews of patients’ experiences with medical facilities and physicians, extracted from social media. This is motivated by the limitations of conventional methods of surveying consumers to assess their satisfaction with the quality of services, which are being replaced by automatic processing of text data from social media. This approach enables to get more objective results due to the increased representativeness and independence of the sample of service consumers. The authors have tested machine learning methods using various neural network architectures. A hybrid method was developed to classify text reviews of medical facilities posted by patients on the two most popular physician review websites in Russia. Overall, more than 60,000 reviews were analysed. The main results are as follows: 1) the classification algorithm developed by the authors has a high efficiency, the best result being achieved by the GRU-based architecture (val_accuracy = 0.9271); 2) applying the named entity search method to text messages following their partitioning improved the classification efficiency for each of the classifiers based on artificial neural networks. To further enhance the classification quality, reviews need to be semantically partitioned by target and sentiment and the resulting fragments need to be analysed separately.

Postpartum haemorrhage (PPH) is a significant cause of maternal morbidity and mortality worldwide, particularly in low-resource settings. This study aimed to develop a predictive model for PPH using early risk factors and rank their importance in terms of predictive ability. The dataset was obtained from an observational case-control study in Northern Rwanda. Various statistical models and machine learning techniques were evaluated, including logistic regression, logistic re-gression with elastic-net regularisation, Random Forest, Extremely Randomized Trees, and gradient boosted trees with XGBoost. Random Forest emerged as the best predictive model for PPH across three different train-test data partitions. The important predictors identified in the study were haemoglobin level during labour and maternal age. However, there were differences in PPH risk factor importance in different data partitions, highlighting the need for further investigation. These findings contribute to understanding PPH risk factors, highlight the importance of considering different data partitions and implementing cross-validation in predictive modelling, and emphasise the value of identifying the appropriate prediction model for the application. Effective PPH pre-diction models are essential for improving maternal health outcomes on a global scale. This study provides valuable insights for healthcare providers to develop predictive models for PPH to identify high-risk women and implement targeted interventions

Background: Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a complex and debilitating disease with a significant global prevalence of over 65 million individuals. It affects various systems, including the immune, neurological, gastrointestinal, and circulatory systems. Studies have shown abnormalities in immune cell types, increased inflammatory cytokines, and brain abnormalities. Further research is needed to identify consistent biomarkers and develop targeted therapies. A multidisciplinary approach is essential for diagnosing, treating, and managing this complex disease. The current study aims at employing explainable artificial intelligence (XAI) and machine learning (ML) techniques to identify discriminative metabolites for ME/CFS. Material and Methods: The present study used a metabolomics dataset of CFS patients and healthy controls, including 26 healthy controls and 26 ME/CFS patients aged 22-72. The dataset encapsulated 768 metabolites, classified into nine metabolic super-pathways: amino acids, carbohydrates, cofactors, vitamins, energy, lipids, nucleotides, peptides, and xenobiotics. Random forest-based feature selection and Bayesian Approach based-hyperparameter optimization were implemented on the target data. Four different ML algorithms [Gaussian Naive Bayes (GNB), Gradient Boosting Classifier (GBC), Logistic regression (LR) and Random Forest Classifier (RFC)] were used to classify individuals as ME/CFS patients and healthy individuals. XAI approaches were applied to clinically explain the prediction decisions of the optimum model. Performance evaluation was performed using the indices of accuracy, precision, recall, F1 score, Brier score, and AUC. Results: The metabolomics of C-glycosyltryptophan, oleoylcholine, cortisone, and 3-hydroxydecanoate were determined to be crucial for ME/CFS diagnosis. The RFC learning model outperformed GNB, GBC, and LR in ME/CFS prediction using the 1000 iteration bootstrapping method, achieving 98% accuracy, precision, recall, F1 score, 0.01 Brier score, and 99% AUC. Conclusion: RFC model proposed in this study correctly classified and evaluated ME/CFS patients through the selected biomarker candidate metabolites. The methodology combining ML and XAI can provide a clear interpretation of risk estimation for ME/CFS, helping physicians intuitively understand the impact of key metabolomics features in the model.

Accurate diagnosis for the 400 million people with rare diseases is critical for healthcare decisions, prognosis, understanding disease mechanisms, and identification of treatments. Despite advances in genome sequencing, barriers such as high interpretation costs, diagnostic expertise, throughput associated delays, and uncertain variant classifications persist, with demand exceeding capacity. Many variant classification methods focus narrowly on specific consequences, leading to the use of complex integrative pipelines that often overlook transcript variability and lack prediction transparency. To overcome these limitations, we introduce DITTO. This transparent, transcript-aware machine-learning method demonstrates superior overall performance in accuracy, recall, and precision when compared to existing tools. DITTO is publicly available at https://github.com/uab-cgds-worthey/DITTO

As in many parts of the world, rural-urban, forest-urban interface areas surrounding urban regions expose the natural areas they interact with to a threat of fire risk that can reach various sizes. This risk has been assessed for various regions of the world using many different methods and numerical models so far. Among these, it is seen that machine learning models have successful applications in risk assessment and risk prediction studies. For the fire risk prediction of Istanbul's yet unurbanized regions, but where the city is anticipated to potentially shift, data was collected using the opportunities provided by Geographic Information Systems and Remote Sensing technologies based on fires that occurred between 2000-2021, and the region was examined. Machine learning methods' Random Forest (RF), Extreme Gradient Boosting (XGB), and Light Gradient Boosting (LGB) models were applied for the classification of factors effective in fire. The best result was given by the RF model with 0.93 accuracy, 0.062 F1 score, and 0.753 Area Under Curve (AUC) value. In the classification of factors in the RF model, the grouping between fire-initiating factors and factors effective in spreading is evident, while this distinction is partially noticeable in the other two models.

Partial least square (PLS) regression is a well-known chemometric method used for predictive modelling, especially in the presence of many variables. Although PLS was not initially developed as a technique for classification tasks, scientists have reportedly used this approach successfully for discrimination purposes. Whereas some non-supervised learning approaches including but not limited to PCA, and k-means clustering do well in identifying/understanding grouping and clustering patterns in multidimensional data, they are limited when the end target is discrimination, making PLS a preferable alternative. A total of fertilized 672 chicken egg hyperspectral imaging data, consisting of 336 white eggs and 336 brown eggs were used in this study. Hyperspectral images in the NIR region of 900-1700 nm wavelength range were captured prior to incubation on day 0 and on days 1-4 after incubation. Eggs were candled on incubation day 5 and broken out on day 10 to confirm fertility. While a total number of 312 and 314 eggs were found to be fertile in the brown and white egg batches respectively, total numbers of non-fertile eggs in the same set of batches were 23 and 21 respectively. Spectral information was extracted from a segmented region of interest (ROI) of each hyperspectral image and spectral transmission characteristics were obtained by averaging the spectral information. A moving-thresholding technique was implemented for discrimination based on PLS regression results on the calibration set. With true positive rates (TPR) of up to 100% obtained at selected threshold values of between 0.50-0.85 and on different days of incubation, the results indicated that the proposed PLS technique can accurately discriminate between fertile and non-fertile eggs. The adaptive PLS approach was thereby presented as suitable for handling hyperspectral imaging-based chicken egg fertility data

Lung cancer (LC) stands as the foremost cause of cancer-related fatality rates worldwide. Early diagnosis significantly enhances patient survival rate. Nowadays, low dose computed tomography (LDCT) is widely employed on chest as a tool for large-scale lung cancer screening. Nonetheless, a large amount of chest radiographs causes an onerous burden on radiologists. Some computer-aided diagnostic (CAD) tools can provide insights to medical images for diagnosis and augmenting diagnostic speed. However, due to variation in parameter settings across different patients, substantial discrepancies in image voxels persist. We find different voxel-size causing a compromise between model generalization and diagnostic efficacy. This study investigates the performance disparities of diagnostic models trained on original images and LDCT images reconstructed to different voxel-sizes but isotropic. The application is to differentiate a nodule to be benign or malignant. Using 11 features, a support vector machine (SVM) is trained on LDCT images with an isotropic voxel having side-length of 1.5 mm for 225 patients in house. The result yields a favorable model performance with an accuracy of 0.9596 and an area under the receiver operating characteristic curve (ROC/AUC) of 0.9855. In addition to furnish CAD tools for clinical applications, future research is encouraged to include LDCT images from multi-centers.

Precipitation type is a key parameter used for better retrieval of precipitation characteristics as well as to understand the cloud-convection-precipitation coupling processes. Ice crystals and water droplets inherently exhibit different characteristics in different precipitation regimes (e.g., convection, stratiform), which reflect on satellite remote sensing measurements that help us distinguish them. The Global Precipitation Measurement (GPM) Core Observatory’s Microwave Imager (GMI) and Dual-Frequency Precipitation Radar (DPR) together provide ample information on global precipitation characteristics. As an active sensor, DPR provides an accurate precipitation type assignment, while passive sensors like GMI are traditionally only used for empirical understanding of precipitation regimes. Using collocated precipitation type flags from DPR as the “truth”, this paper employs machine learning (ML) models to train and test the predictability and accuracy of using passive GMI-only observations together with ancillary information from reanalysis and GMI surface emissivity retrieval products. Out of six ML models, four simple ones (Support Vector Machine, Neural Network, Random Forest, and Gradient Boosting) and the 1-D convolutional neural network (CNN) model are identified to produce 90% - 94% prediction accuracy globally for 5 types of precipitation (convective, stratiform, mixture, no precipitation, and other precipitation), which is much more robust than previous similar effort. One novelty of this work is to introduce data augmentation (subsampling and bootstrapping) to handle extremely unbalanced samples in each category. Careful evaluation of Impact matrices demonstrate that polarization difference (PD) and surface emissivity at high-frequency channels dominate the decision process, which are consistent with the physical understanding of polarized microwave radiative transfer over different surface types, as well as in snow and liquid clouds with different microphysical properties. Furthermore, the view-angle dependency artifact that DPR precipitation flag bears with does not propagate into the conical-viewing GMI retrievals. This work provides a new and promising way for future physics-based ML retrieval algorithm development.

Herein, we examined the effects of weather parameters and individual clinicodemographic risk factors on hemorrhagic stroke (HS) incidence, severity on admission, and disability at discharge in a harsh desert climate. In a retrospective design we studied patients admitted to a stroke unit in Arab Emirates in 2016-2019. With a distributed lag nonlinear model we explored immediate and delayed effects of weather on stroke incidence. Supervised machine learning was used to build models predictive of scores in NIHSS and mRS scales. We assessed model performance by calculating ROC AUC, F1 scores, specificity, and sensitivity. HS risk increased after a significant change in any weather parameter. The risk was reduced below baseline after 1-week adjustment to environmental changes. Climate input from the previous 3 days yielded reliable classification models. Humidex was a better predictor of severity than air temperature, whereas body mass index, age, and time of day at stroke onset were the strongest clinical predictors. Weather parameters before admission were stronger predictors of disease severity than individual clinical factors. We addressed limitations of previous studies by analyzing a full set of climate parameters: ambient temperature, relative humidity, humidex, atmospheric pressure, and wind speed. Predictive models may help to optimize patient management and develop preventative strategies.

Using machine learning or deep learning to solve the problem of garbage recognition and classification is an important application in computer vision, but due to the incomplete establishment of garbage datasets and the poor performance of complex network models on smart terminal devices, the existing garbage classification models The effect is not good.This paper presents a waste classification and identification method base on transfer learning and lightweight neural network. By migrating the lightweight neural network MobileNetV2 and rebuild it, The reconstructed network is used for feature extraction, and the extracted features are introduced into the SVM to realize the identification of 6 types of garbage. The model was trained and verified by using 2527 pieces of garbage labeled data in the TrashNet dataset, which ultimately resulted in a classification accuracy of 98.4% of the method, which proves that the method can effectively improve the classification accuracy and time and overcome the problem of weak data and less labeling. The over-fitting phenomenon encountered by small data sets in deep learning makes the model robust.

Several factors impact the durability of concrete bridge decks, including traffic loads, fatigue, temperature changes, environmental stress, and maintenance activities. Detecting problems such as corrosion, delamination, or concrete degradation early on can lower maintenance costs. Nondestructive evaluation (NDE) techniques can detect these issues at early stages. Each NDE method, meanwhile, has limitations that reduce the accuracy of the assessment. In this study, multiple NDE technologies were combined with machine learning algorithms to improve the interpretation of half-cell potential (HCP) and electrical resistivity (ER) measurements. A parametric study was performed to analyze the influence of five parameters on HCP and ER measurements, such as degree of saturation, corrosion length, delamination depth, concrete cover, and moisture condition of delamination. The results were obtained through finite element simulations and used to build two machine learning algorithms, a classification algorithm and a regression algorithm, based on the Random Forest methodology. The algorithms were tested using data collected from a bridge deck in the BEAST® facility. Both machine learning algorithms were effective in improving the interpretation of ER and HCP measurements using data from multiple NDE technologies.

A chronic activated pro-inflammatory cytokine network (“inflamm-aging”) may amplify the neurodegenerative effects of a fall induced brain trauma in geriatric subjects. Our research aimed to evaluate how a trained machine learning algorithm may predict recent antecedent falls based only on specific serologic cytokines network analysis and how the consequences of these falls can be substantiated on standard head MRIs. All 279 subjects included in our study were selected from the ADNI1 dataset and all had a mild cognitive impairment diagnostic at the ADNI1 study baseline. A “train group” was built and included 14 subjects with a history of a recent, simple, standing-level fall. These were carefully matched with 14 similar subjects without any antecedent trauma. The “test group” included 251 subjects, all without any history of recent fall. The machine learning algorithm (classic C4.5 decision tree) was trained to detect a pattern of variation in 23 clinically relevant cytokines in relation with an antecedent fall. Changes in five cytokines (matrix metalloproteinase-7, eotaxin-1, interleukin-3, interleukin-8 and matrix metalloproteinase-9) were used for fall prediction in the “test” group. Once trained, the algorithm predicted a recent fall in 119 cases from the test group. The mean brain ventricular volume that was significantly different between fall/non-fall subgroups (41645.5±10337.2 vs 27127.3±6749.4 mm3, p=0.005) remained significant in the test group, after prediction between (41544.24±17343.4 vs 34553.5±10543.2 mm3, p=0.042). The hippocampus mean volume was also significantly different between in the test group (6297.3±1080.1 vs 6745.9±1123.7, p=0.0015). A significant brain ventricular difference was observed in the “65<y.o.” subgroup (p=0.04). If confirmed by larger prospective studies, our findings may increase the precision of the neuro-cognitive assessments in geriatric subjects.

Neutral facial expression recognition is of great importance in various domains and applications. This study introduces a data-centric approach for neutral facial expression recognition, presenting a comprehensive study that explores different methodologies, techniques, and challenges in the field to foster a deeper understanding. The results show that data augmentation plays a crucial role in improving dataset performance. Additionally, the study investigates different model architectures and training techniques to identify the most effective approach, with the InceptionV3 model achieving the highest accuracy of 72%. Furthermore, the research examines the influence of preprocessing methods on the performance of both InceptionV3 and a simplified CNN model. Interestingly, the results indicate that preprocessing techniques positively affect the performance of the simpler CNN model but negatively impact the InceptionV3 model. The implemented system, used to evaluate the findings, demonstrates promising results, correctly classifying 77% of neutral expressions. However, there are still areas for improvement. Creating a specialized dataset that includes both neutral and non-neutral expressions would greatly enhance the accuracy of the system. By addressing limitations and implementing suggested improvements, neutral facial expression recognition can be significantly enhanced, leading to more effective and accurate results.

Urban spatial perception critically influences human behavior and emotional reactions, emphasizing the necessity of aligning urban spaces with human needs for enhanced urban living. However, functionality-based categorization of urban architecture is prone to biases, stemming from disparities between objective mapping and subjective perception. These biases can result in urban planning and designs that fail to cater adequately to the needs and preferences of city residents, negatively impacting their quality of life and the city's overall functionality. In this study, we apply machine learning to elucidate these biases within urban spatial perception research, utilizing a three-step methodology: objective mapping, subjective perception analysis, and perception deviation assessment. Our findings reveal that machine learning can expose hidden patterns within this research field, bearing substantial implications for urban planning and design. Of particular note, the study revealed significant discrepancies in the distribution centroids between commercial buildings and residential or public buildings. This result illuminates the spatial organization characteristics of urban architectural functions, serving as a valuable reference for urban planning and development. Moreover, it uncovers the advantages and disadvantages of different data sources and techniques in interpreting urban spatial perception, paving the way for a more comprehensive understanding of the subject. These findings underscore the importance of integrating both objective mapping and subjective perspectives in urban architectural functionality classification.

In the field of pavement engineering, the determination of the mechanical characteristics is one of the essential process for reliable material design and highway sustainability. Early determination of mechanical characteristics of pavement is highly essential for road and highway construction and maintenance. Tensile strength (TS), compressive strength (CS) and flexural strength (FS) of roller compacted concrete pavement (RCCP) are very crucial characteristics as they are necessitated for many data from mixture proportions as input variables. In this research, the classification-based regression models named Random Forest (RF), M5rule model tree (M5rule), M5prime model tree (M5p) and Chi-square Automatic Interaction Detection (CHAID) are developed for simulation of the mechanical characteristics of RCCP. A comprehensive and reliable dataset comprising 621, 326 and 290 data records for CS, TS and FS experimental cases extracted from several open sources over the literature. The mechanical properties are developed based on influential inputs combination that processed using Principle Component Analysis (PCA). The applied PCA method as feature selection is specified that volumetric/weighted content forms of experimental variables (e.g., coarse aggregate, fine aggregate, supplementary cementitious materials, water and binder) and specimens’ age are the most effective inputs to generate the better performances. Several statistical metrics are measured to evaluate proposed classification-based regression models. RF model revealed an optimistic classification capacity of the CS, TS and FS prediction of the RCCP in comparison with the CHAID, M5rule, and M5p models. The research is extended for the results verification using Monte-carlo model for the uncertainty and sensitivity of variables importance analysis. Overall, the proposed methodology indicated a reliable soft computing model that can be implemented for the material engineering construction and design.

CNN-based methods have dominated the field of aerial scene classification for the past few years. While achieving remarkable success, CNN-based methods suffer from excessive parameters and notoriously rely on large amounts of training data. In this work, we introduce few-shot learning to the aerial scene classification problem. Few-shot learning aims to learn a model on base-set that can quickly adapt to unseen categories in novel-set, using only a few labeled samples. To this end, we proposed a meta-learning method for few-shot classification of aerial scene images. First, we train a feature extractor on all base categories to learn a representation of inputs. Then in the meta-training stage, the classifier is optimized in the metric space by cosine distance with a learnable scale parameter. At last, in the meta-testing stage, the query sample in the unseen category is predicted by the adapted classifier given a few support samples. We conduct extensive experiments on two challenging datasets: NWPU-RESISC45 and RSD46-WHU. The experimental results show that our method yields state-of-the-art performance. Furthermore, several ablation experiments are conducted to investigate the effects of dataset scale, the impact of different metrics and the number of support shots; the experiment results confirm that our model is specifically effective in few-shot settings.

In an era of rapid technological improvements, state-of-the-art methodologies and tools dedicated to protecting and promoting our cultural heritage should be developed and extensively employed in the contemporary built environment and lifestyle. At the same time, sustainability principles underline the importance of the continuous use of historic or vernacular buildings as part of the building stock of our society. Adopting a holistic, integrated, multi-disciplinary strategy can bridge technological innovation with conserving and restoring heritage buildings. The paper presents ongoing research and results of the application of Machine Learning methods for the remote monitoring of the built environment of the historic cluster in Cypriot cities. This study is part of an integrated, multi-scale, and multi-discipline study of heritage buildings towards the creation of an online HBIM platform for urban monitoring.

Accurate transfer learning of clinical outcomes, e.g., of the effects and side effects of drugs or other interventions, from one cellular context to another (in-vitro versus ex-vivo versus in-vivo, or across tissues), between cell-types, developmental stages, omics modalities or species, is considered tremendously useful. Ultimately, it may avoid most drug development failing in translation, despite large investments in the preclinical stages, which includes animal experiments requiring careful justification. Thus, when transferring a prediction task from a source (model) domain to a target domain, what counts is the high quality of the predictions in the target domain, requiring molecular states or processes common to both source and target that can be learned by the predictor, reflected by latent variables. These latent variables may form a compendium of knowledge that is learned in the source, to enable predictions in the target; usually, there are few, if any, labeled target training samples to learn from. Transductive learning then refers to the learning of the predictor in the source domain, transferring its outcome label calculations to the target domain, considering the same task. Inductive learning considers cases where the target predictor is performing a different yet related task as compared to the source predictor, making some labeled target data necessary. Often, there is also a need to first map the variables in the input/feature spaces (e.g. of gene names to orthologs) and/or the variables in the output/outcome spaces (e.g. by matching of labels). Transfer across omics modalities also requires that the molecular information flow connecting these modalities is sufficiently conserved. Only one of the methods for transfer learning we reviewed offers an assessment of input data, suggesting that transfer learning is unreliable in certain cases. Moreover, source domains feature their very own particularities, and transfer learning should consider these, e.g., as differences in pharmacokinetics, drug clearance or the microenvironment. In light of these general considerations, we here discuss and juxtapose various recent transfer learning approaches, specifically designed (or at least adaptable) to predict clinical (human in-vivo) outcomes based on molecular data, towards finding the right tool for a given task, and paving the way for a comprehensive and systematic comparison of the suitability and accuracy of transfer learning of clinical outcomes.

Chronic venous disease (CVD) occurs in a substantial proportion of the world's population. If the onset of CVD looks like a cosmetic defect, then over time, it can develop into serious problems that require surgical intervention. The aim of the work is to use deep learning (DL) methods for automatic classification of the stage of CVD for self-diagnosis of a patient by using the image of the patient’s legs. The required for DL algorithms images of legs with CVD were obtained by using Internet Data Mining. For images preprocessing, the binary classification problem “legs - no legs” was solved based on Resnet50 with accuracy 0.998. The application of this filter made it possible to collect a data set of 11,118 good quality leg images with various stages of CVD. For classification of various stages of CVD according to CEAP classification, the multi classification problem was set and resolved by using two neural networks with completely different architecture - Resnet50 and DeiT. The model based on DeiT without any tuning shows better results than the model based on Resnet50 (precision = 0.770 (DeiT) and 0.615 (Resnet50)). To demonstrate the results of the work, a telegram bot was developed, in which fully functioning DL algorithms are implemented. This bot allows evaluating the condition of the patient's legs with a fairly good accuracy for the CVD classification.

As the Autonomous Vehicle (AV) industry is rapidly advancing, classification of non-motorized (vulnerable) road users (VRUs) becomes essential to ensure their safety and to smooth operation of road applications. The typical practice of non-motorized road users’ classification usually takes numerous training time and ignores the temporal evolution and behavior of the signal. In this research effort, we attempt to detect VRUs with high accuracy be proposing a novel framework that includes using Deep Transfer Learning, which saves training time and cost, to classify images constructed from Recurrence Quantification Analysis (RQA) that reflect the temporal dynamics and behavior of the signal. Recurrence Plots (RPs) were constructed from low-power smartphone sensors without using GPS data. The resulted RPs were used as inputs for different pre-trained Convolutional Neural Network (CNN) classifiers including constructing 227×227 images to be used for AlexNet and SqueezeNet; and constructing 224×224 images to be used for VGG16 and VGG19. Results show that the classification accuracy of Convolutional Neural Network Transfer Learning (CNN-TL) reaches 98.70%, 98.62%, 98.71%, and 98.71% for AlexNet, SqueezeNet, VGG16, and VGG19, respectively. The results of the proposed framework outperform other results in the literature (to the best of our knowledge) and show that using CNN-TL is promising for VRUs classification. Because of its relative straightforwardness, ability to be generalized and transferred, and potential high accuracy, we anticipate that this framework might be able to solve various problems related to signal classification.

Keywords: Electrical power systems, Support vector machines, Random Forest, Machine learning, Wavelet transform, Transmission lines fault, Electrical power quality, Short circuit, Classification of faults, Localization of faults, Decision trees, Ensemble learning, K-nearest neighbors.

This paper uses the numerical results of surveys sent to Huelva’s (Andalusia, Spain) households to determine the degree of knowledge they have about the urban water cycle, needs, values, and attitudes regarding water in an intermediary city with low water stress. In previous research, we achieved three different households’ clusters. The first one grouped households with high knowledge of the integral water cycle and a positive attitude to smart devices at home. The second cluster described households with low knowledge of the integral water cycle and high sensitivity to price. The third one showed average knowledge and predisposition to have a closer relationship with the water company. This paper continues with this research line, applying Classification and Regression Trees (CART) to determine which hierarchy of variables/factors/ independent components obtained from the surveys are the decisive ones to predict the range of household water consumption in Huelva. Positive attitudes towards improved cleaning habits for personal or household purposes are the highest hierarchy component to predict the water consumption range. Second in the hierarchy, the variable Knowledge Global Score about the integral urban water cycle, associated with water literacy, also contributes to predicting the water consumption range. Together with the three clusters obtained previously, these results will allow us to design water demand management strategies (WDM) fit for purpose that enable Huelva’s households to use water more efficiently.

The paradigm of leveraging code as a dataset has recently gained traction, offering innovative solutions in domains such as automated commit message generation, pull request description automation, and program repair mechanisms. Consider the challenge in generating commit messages: traditional methods treat source code as a mere token sequence, applying neural machine translation models. This approach, however, overlooks the critical syntactic structures inherent in programming languages, which could offer deeper insights and improved accuracy. Building upon prior research, specifically the Code2Seq framework, which utilized Abstract Syntax Tree (AST) structural data for source code representation to automate method name generation, this paper extends and refines this concept. We introduce "CSR", a novel methodology adapted to represent code edits effectively. This paper investigates the impact of employing syntactic structure, focusing on the classification of code edits. Drawing inspiration from Code2Seq, "CSR" utilizes AST's structural properties, particularly the paths connecting leaf nodes, to enhance the task of code edit classification. This approach is rigorously evaluated on two distinct datasets, comprising fine-grained syntactic edits. Our comprehensive experiments reveal that incorporating syntactic structures does not significantly outperform simpler methodologies. While methods like Code2Seq and our proposed "CSR" show potential, our findings highlight that there is considerable scope for improvement and refinement before such techniques can be universally applied for learning representations of code edits. We anticipate that our findings will spark further research in this field, paving the way for more effective use of syntactic structures in code representation.

Hyperspectral image (HSI) classification is a mechanism of analyzing differentiated land cover in remotely sensed hyperspectral images. In the last two decades, a number of different types of classification algorithms have been proposed for classifying hyperspectral data. These algorithms include supervised as well as unsupervised methods. Each of these algorithms has its own limitations. In this research, three different types of unsupervised classification methods are used to classify different datasets i-e Pavia Center, Pavia University, Cuprite, Moffett Field. The main objective is to assess the performance of all three classifiers K-Means, Spectral Matching, and Abundance Mapping, and observing their applicability on different datasets. This research also includes spectral feature extraction for hyperspectral datasets.

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%.

Computer vision methods have shown to be effective in classifying garbage into recycling categories for waste processing, existing methods are costly, imprecise, and unclear. To tackle this issue, we introduce MWaste, a mobile application that uses computer vision and deep learning techniques to classify waste materials as trash, plastic, paper, metal, glass or cardboard. Its effectiveness was tested on various neural network architectures and real world images, achieving an average precision of 92% on the test set. This app can help combat climate change by enabling efficient waste processing and reducing the generation of greenhouse gases caused by incorrect waste disposal.

While visual assessment is the standard technique for burn evaluation, computer-aided diagnosis is increasingly sought due to high number of incidences globally. Patients are increasingly facing challenges which are not limited to shortage of experienced clinicians, lack of accessibility to healthcare facilities, and high diagnostic cost. Certain number of studies were proposed in discriminating burn and healthy skin using machine learning leaving a huge and important gap unaddressed; whether burns and related skin injuries can be effectively discriminated using machine learning techniques. Therefore, we specifically use pre-trained deep learning models due to deficient dataset to train a new model from scratch. Experiments were extensively conducted using three state-of-the-art pre-trained deep learning models that includes ResNet50, ResNet101 and ResNet152 for image patterns extraction via two transfer learning strategies: fine-tuning approach where dense and classification layers were modified and trained with features extracted by base layers, and in the second approach support vector machine (SVM) was used to replace top-layers of the pre-trained models, trained using off-the-shelf features from the base layers. Our proposed approach records near perfect classification accuracy of approximately 99.9%.

Effective healthcare relies on accurate and timely diagnosis; however, obtaining large amounts of training data while maintaining patient privacy remains challenging. This study introduces a novel approach utilizing federated learning (FL) and a cross-device multi-modal model for clin-ical event classification using vital signs data. Our architecture leverages FL to train machine learning models, including Random Forest, AdaBoost, and SGD ensemble model, on vital signs data from a diverse clientele at a Boston hospital (MIMIC-IV dataset). The FL structure preserves patient privacy by training directly on each client's device without transferring sensitive data. The study demonstrates the potential of FL in privacy-preserving clinical event classification, achieving an impressive accuracy of 98.9%. These findings underscore the significance of FL and cross-device ensemble technology in healthcare applications, enabling the analysis of large amounts of sensitive patient data while safeguarding privacy.

Object classification is a crucial task in deep learning, which involves the identification and categorization of objects in images or videos. Although humans can easily recognize common objects, such as cars, animals, or plants, performing this task on a large scale can be time-consuming and error-prone. Therefore, automating this process using neural networks can save time and effort while achieving higher accuracy. Our study focuses on the classification step of human chromosome karyotyping, an important medical procedure that helps diagnose genetic disorders. Traditionally, this task is performed manually by expert cytologists, which is a time-consuming process that requires specialized medical skills. Therefore, automating it through deep learning can be immensely useful. To accomplish this, we implemented and adapted existing preprocessing and data augmentation techniques to prepare the chromosome images for classification. We used ResNet-50 convolutional neural networks and an ensemble approach to classify the chromosomes, obtaining state-of-the-art performance in the tested dataset.

Commercial banks usually classify customers according to their credit reports when making loans. In this study, we put our focus on classifying customers based on their credit reports from the People's Bank of China (PBC). Since there are no target labels of users in the credit report of the People's Bank of China, we put forward the fuzzy clustering method for the initial label, and then Construct ant colony search to optimize intelligent recognition. Finally, this study uses SVM, BP neural network, and random forest to classify users and compare their results. The research re-sults indicate that using ant colony clustering algorithm and random forest for classification is the most effective method with the PBC credit reports.

The task of unsupervised anomalous sound detection (ASD) is challenging for detecting anomalous sounds from a large audio database without any annotated anomalous training data. Many unsupervised methods were proposed, but previous works have confirmed that the classification-based models far exceeds the unsupervised models in ASD. In this paper, we adopt two classification-based anomaly detection models: (1) Outlier classifier is to distinguish anomalous sounds or outliers from the normal; (2) ID classifier identifies anomalies using both the confidence of classification and the similarity of hidden embeddings. We conduct experiments in task 2 of DCASE 2020 challenge, and our ensemble method achieves an averaged area under the curve (AUC) of 95.82% and averaged partial AUC (pAUC) of 92.32%, which outperforms the state-of-the-art models.

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.

A pixel-based segmentation method was demonstrated to be confounded by developmental stage in estimation of flowering of mango. Categorization of panicles into three developmental stages was undertaken with a single and a two-stage deep learning framework (YOLO and R2CNN), using either upright or rotated bounding boxes. For a validation image set and for total panicle count, the models MangoYOLO(-upright), MangoYOLO-rotated, YOLOv3-rotated, R2CNN(-rotated) and R2CNN-upright achieved: (i) RMSEs of 25.6, 16.0, 15.4, 25.8 and 32.3 panicles per tree image, (ii) Mean average precision (mAP) scores of 72.2, 69.1, 65.0, 62.5 and 70.9% and (iii) weighted F1-scores of 76.5, 76.1, 74.9, 74.0 and 82.0, respectively. For a test set of images involving a different orchard and cultivar and use of a different camera, the R2 for machine vision to human count of panicles per tree was 0.86, 0.80, 0.83, 0.81 and 0.76 for the same models, respectively. Thus, models generalised well, but with no consistent benefit from use of rotated over upright bounding boxes. While the YOLOv3-rotated model was superior in terms of total panicle count, the R2CNN-upright model was more accurate for panicle stage classification. To demonstrate practical application, panicle counts were made weekly for an orchard of 994 trees, with a peak detection routine applied to document multiple flowering events.

Increasingly measured data in the context of smart cities can be used to develop new and innovative business models to increase efficiency and the value of life. A time-series classification algorithm can support to automatize many different processes such as forecasting services. In order to ensure data security and privacy, Federated Learning trains a global model collaboratively on multiple clients. Having different data-distributions and data-quantities across participating clients, neural networks suffer from slow convergence and overfitting. Based on different data-distributions, data-quantities and number of clients, we develop and evaluate different data-clustering strategies to update global model weights in comparison to the state of the art. We use public time-series data, generate various synthetic datasets and train a Relational-Regularized Autoencoder for classification purposes. Our results show an improvement of model performance concerning generalization.

In videos, the human's actions are of three-dimensional (3D) signals. These videos investigate the spatiotemporal knowledge of human behavior. The promising ability is investigated using 3D convolution neural networks (CNNs). The 3D CNNs have not yet achieved high output for their well-established two-dimensional (2D) equivalents in still photographs. Board 3D Convolutional Memory and Spatiotemporal fusion face training difficulty preventing 3D CNN from accomplishing remarkable evaluation. In this paper, we implement Hybrid Deep Learning Architecture that combines STIP and 3D CNN features to enhance the performance of 3D videos effectively. After implementation, the more detailed and deeper charting for training in each circle of space-time fusion. The training model further enhances the results after handling complicated evaluations of models. The video classification model is used in this implemented model. Intelligent 3D Network Protocol for Multimedia Data Classification using Deep Learning is introduced to further understand space-time association in human endeavors. In the implementation of the result, the well-known dataset, i.e., UCF101 to, evaluates the performance of the proposed hybrid technique. The results beat the proposed hybrid technique that substantially beats the initial 3D CNNs. The results are compared with state-of-the-art frameworks from literature for action recognition on UCF101 with an accuracy of 95%.

Data sets for medical images are generally imbalanced and limited in sample size because of high data collection costs, time-consuming annotations, and patient privacy concerns. The training of deep neural network classification models on these data sets to improve the generalization ability does not produce the desired results for classifying the medical condition accurately and often overfit the data on the majority of class samples. To address the issue, we propose a framework for improving the classification performance metrics of deep neural network classification models using transfer learning: pre-trained models, such as Xception, InceptionResNet, DenseNet along with the Generative Adversarial Network (GAN) – based data augmentation. Then, we trained the network by combining traditional data augmentation techniques, such as randomly flipping the image left to right and GAN-based data augmentation, and then fine-tuned the hyper-parameters of the transfer learning models, such as the learning rate, batch size, and the number of epochs. With these configurations, the Xception model outperformed all other pre-trained models achieving a test accuracy of 98.7%, the precision of 99%, recall of 99.3%, f1-score of 99.1%, receiver operating characteristic (ROC) - area under the curve (AUC) of 98.2%.

Evolution remains an incessant process in viruses, allowing them to elude host immune response and induce severe diseases, impacting the diagnostic and vaccine effectiveness. Predicting emerging viral genomes is crucial, particularly in diseases like dengue, where viruses disrupt host cells, leading to fatal outcomes. Deep learning has been applied to predict dengue fever cases; there has been relatively less emphasis on its significance in forecasting emerging Dengue Virus (DENV) serotype. While Recurrent Neural Networks (RNN) were originally designed for modeling temporal sequences, our proposed DL-DVE generative and classification model, trained on complete genome data of DENV, transcends traditional approaches by learning semantic relationships between nucleotides in a continuous vector space instead of representing contextual meaning of nucleotide characters. Leveraging 2000 publicly available DENV complete genome sequences, our Long Short-Term Memory (LSTM) based generative and Feedforward Neural Network (FNN) based classification DL-DVE model showcases proficiency in learning intricate patterns and generating sequences for emerging serotype of DENV. The generative model showed accuracy of 93% and the classification model provided insight into the specific serotype label, corroborated by BLAST search verification. Evaluation metrics such as ROC-AUC value 0.818, accuracy, precision, recall and F1 score all to be around 99.00%, demonstrated the classification model’s reliability.  Our model classified the generated sequences as DENV-4, exhibiting 65.99% similarity to DENV-4 and around 63-65% similarity with other serotypes, indicating notable distinction from other serotypes. Moreover, the intra-serotype divergence of sequences with a minimum 90% similarity underscored their uniqueness. We analyzed the conserved motifs in the genome through MEME Suite (version 5.5.5). Our research strives to contribute to the ongoing fight against the Dengue virus by offering predictive insights into its genomic evolution. Looking ahead, proactive predictive modeling before mutations occur holds potential for guiding vaccine design and diagnostic kit development.

Machine learning applications often need large amounts of training data to perform well. Whereas unlabeled data can be easily gathered, the labeling process is difficult, time-consuming, or expensive in most applications. Active learning can help solve this problem by querying labels for those data points that will improve the performance the most. Thereby, the goal is that the learning algorithm performs sufficiently well with fewer labels. We provide a library called scikit-activeml that covers the most relevant query strategies and implements tools to work with partially labeled data. It is programmed in Python and builds on top of scikit-learn.

In the field of computer vision, large-scale image classification tasks are both important and highly challenging. With the ongoing advances in deep learning and optical character recognition (OCR) technologies, neural networks designed to perform large-scale classification play an essential role in facilitating OCR systems. In this study, we developed an automatic OCR system designed to identify up to 13,070 large-scale printed Chinese characters by using deep learning neural networks and fine-tuning techniques. The proposed framework comprises four components, including training dataset synthesis and background simulation, image preprocessing and data augmentation, the process of training the model, and transfer learning. The training data synthesis procedure is composed of a character font generation step and a background simulation process. Three background models are proposed to simulate the factors of the background noise and anti-counterfeiting patterns on ID cards. To expand the diversity of the synthesized training dataset, rotation and zooming data augmentation are applied. A massive dataset comprising more than 19.6 million images was thus created to accommodate the variations in the input images and improve the learning capacity of the CNN model. Subsequently, we modified the GoogLeNet neural architecture by replacing the FC layer with a global average pooling layer to avoid overfitting caused by a massive amount of training data. Consequently, the number of model parameters was reduced. Finally, we employed the transfer learning technique to further refine the CNN model using a small number of real data samples. Experimental results show that the overall recognition performance of the proposed approach is significantly better than that of prior methods and thus demonstrate the effectiveness of proposed framework, which exhibited a recognition accuracy as high as 99.39% on the constructed real ID card dataset.

Deep learning heavily relies on statistical correlations to drive artificial intelligence (AI) innovations, particularly in computer vision applications like autonomous driving and robotics. However, despite providing a solid foundation for deep learning, these statistical correlations can be vulnerable to unforeseen and uncontrolled factors. The lack of prior knowledge guidance can result in spurious correlations, introducing confounding factors and affecting the model's robustness. To address this challenge, recent research efforts have focused on integrating causal theory into deep learning methodologies. By modelling the inherent and unbiased causal structure, causal theory can potentially mitigate the impact of spurious correlations effectively. Hence, this paper explores the basics of causal methodologies in image classification.

In August 2020, the World Health Assembly launched a global initiative to eliminate cervical cancer by 2030, setting three primary targets. One key goal is to achieve a 70% screening coverage rate for cervical cancer, primarily relying on the precise analysis of Papanicolaou (Pap) or digital Pap smears. However, the responsibility of reviewing Pap smear samples to identify potentially cancerous cells primarily falls on pathologists, a task known to be exceptionally challenging and time-consuming. This paper proposes a solution to address the shortage of pathologists for cervical cancer screening. It leverages the OpenAI-GYM API to create a deep reinforcement learning environment utilizing liquid-based Pap smear images. By employing the Proximal Policy Optimization algorithm, autonomous agents navigate Pap smear images, identifying cells with the aid of rewards, penalties, and accumulated experiences. Furthermore, the use of a pre-trained convolutional neuronal network like Res-Net54 enhances the classification of detected cells based on their potential for malignancy. The ultimate goal of this study is to develop a highly efficient, automated Papanicolaou analysis system, ultimately reducing the need for human intervention in regions with limited pathologists.

Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify the physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the proposed models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from 8 volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides the accuracy performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. Our results indicate that the proposed method will provide the medical doctors and trainers a promising way to precisely track and understand a patient’s physical activities for better treatment.

Our daily activities hinge on the flexibility of our fingers, and a fractured finger can significantly disrupt these routines. The finger bones enable us to bend and fold the fingers and thumb to pick up or grasp objects and do all of our daily activities. A broken finger can cause adverse effects on our daily life activities. It is important to treat broken finger as soon as possible. Swift and precise treatment begins with capturing multiple X-rays, followed by the critical step of fracture detection in these images. Relying on the naked eye for this task carries the risk of overlooking small fractures. To address this issue an automated diagnoses of fractured fingers from images is required for which the current research employs advanced deep learning models—ResNet34, ResNet50, ResNet101, ResNet152, VGG-16, and VGG-19—to classify finger images from the Musculoskeletal Radiographs (MURA) dataset as either fractured or non-fractured. The results emphasize the consistently strong performance of ResNet models, attaining an impressive accuracy of 81.9%. This surpasses VGG models by 3.4% and establishes ResNet as the new benchmark for state-of-the-art accuracy.

Recent neuroscience research results show that the nerve information in the brain is not only encoded by the spatial information. Spiking neural network based on pulse frequency coding plays a very important role in dealing with the problem of brain signal, especially complicated space-time information. In this paper, an unsupervised learning algorithm for bilayer feedforward spiking neural networks based on spike-timing dependent plasticity (STDP) competitiveness is proposed and applied to SAR image classification on MSTAR for the first time. The SNN learns autonomously from the input value without any labeled signal and the overall classification accuracy of SAR targets reached 80.8%. The experimental results show that the algorithm adopts the synaptic neurons and network structure with stronger biological rationality, and has the ability to classify targets on SAR image. Meanwhile, the feature map extraction ability of neurons is visualized by the generative property of SNN, which is a beneficial attempt to apply the brain-like neural network into SAR image interpretation.

The research of object classification and part segmentation is a hot topic in computer vision. A considerable number of studies have been carried out about deep learning on 3D point clouds. However,it is challenging to achieve effective feature learning due to sparsity of point clouds. Recently, a variety of Transformers have been adopted to improve point cloud processing and display great potential. Nevertheless, large numbers of Transformer layers tend to incur huge computational and memory costs. PointNet++ is one of the most influential neural architectures for point cloud understanding. Although the accuracy of PointNet++ has been largely surpassed by recent networks, this does not mean that PointNet++ has no potential.Thus, this paper offer two major contributions that significantly improve PointNet++ performance. Firstly, we introduce a novel contextual feature extraction (CFE) block that significantly enhances the feature extraction capabilities of PointNet++ networks. Secondly, to further enhance feature fusion, we seamlessly integrate a mutual learning (ML) block into the network architecture. By embedding these two innovative blocks within each layer of the network, we not only enrich the network's functionality but also impart it with greater robustness and adaptability. The specific experiments were conducted on the S3DIS (6-fold cross-validation) and Modelnet40 datasets with 86.5% and 92.7% accuracy, respectively, which proved that our model is comparable or even better than most existing methods for classification and segmentation tasks.

The objectives of this paper are to investigate the tradeoffs between a physically constrained neural network and a deep, convolutional neural network and to design a combined ML approach ("VarioCNN"). Our solution is provided in the framework of a cyberinfrastructure that includes a newly designed ML software, GEOCLASS-image, modern high-resolution satellite image data sets (Maxar WorldView data) and instructions/descriptions that may facilitate solving similar spatial classification problems. Combining the advantages of the physically-driven connectionist-geostatistical classification method with those of an efficient CNN, VarioCNN provides a means for rapid and efficient extraction of complex geophysical information from submeter resolution satellite imagery. A retraining loop overcomes the difficulties of creating a labeled training data set. Computational analyses and developments are centered on a specific, but generalizable, geophysical problem: The classification of crevasse types that form during the surge of a glacier system. A surge is a glacial catastrophe, an acceleration of a glacier to typically 100-200 times its normal velocity. GEOCLASS-image is applied to study the current (2016-2024) surge in the Negribreen Glacier System, Svalbard. The geophysical result is a description of the structural evolution and expansion of the surge, based on crevasse types that capture ice deformation in six simplified classes.

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.