Pseudo-Multiview Learning Using Subjective Logic for Enhanced Classification Accuracy

Today data's are stored in relation structures.In usual approach to mine these data, we often use to join several relations to form a single relation using foreign key links, which is known as flatten.Flatten may cause troubles such as time consuming, data redundancy and statistical skew on data.Hence, the critical issues arise that how to mine data directly on numerous relations.The solution of the given issue is the approach called multi-relational data mining (MRDM).Other issues are irrelevant or redundant attributes in a relation may not make contribution to classification accuracy.Thus, feature selection is an essential data pre-processing step in multi-relational data mining.By filtering out irrelevant or redundant features from relations for data mining, we improve classification accuracy, achieve good time performance, and improve comprehensibility of the models.We had proposed the entropy based feature selection method for Multi-relational Naïve Bayesian Classifier.We have use method InfoDist and Pearson's Correlation parameters, which will be used to filter out irrelevant and redundant features from the multi-relational database and will enhance classification accuracy.We analyzed our algorithm over PKDD financial dataset and achieved the better accuracy compare to the existing features selection methods.

Background/Objectives: Multimodal data fusion is increasingly applied in neuroinformatics to integrate heterogeneous sources of information. However, the optimal strategies for combining modalities with markedly different dimensionality, scale, and noise characteristics remain unclear. To our knowledge, this is among the first systematic and controlled benchmarks explicitly disentangling the effects of fusion strategy and feature scaling within a unified deep learning framework. Methods: Using data from 747 healthy participants from the Human Connectome Project, we evaluated multiple fusion paradigms-including early fusion, attention-based fusion, subspace-based fusion, and graph-based fusion-within a unified and reproducible framework. Importantly, we assessed how different feature scaling techniques (Standard, Min-Max, and Robust scaling) interact with fusion strategies and influence model performance. Biological sex was used as a controlled benchmark task to focus on methodological insights rather than task-specific optimization. Results: Early feature-level fusion consistently achieved the highest classification performance across all evaluated configurations. In particular, direct concatenation of cognitive and neuroimaging features combined with Standard Scaling yielded the best results (AUC-ROC = 0.96 (0.95-0.96)), outperforming unimodal baselines as well as intermediate and late fusion strategies. Conclusions: This systematic benchmark demonstrates that multimodal deep learning performance in neuroscience is driven primarily by the interaction between fusion strategy and feature scaling rather than by architectural complexity alone. By explicitly disentangling the effects of fusion level and preprocessing within a unified framework, this study provides practical methodological guidance for the design, evaluation, and reproducible deployment of multimodal deep learning models in neuroscience.

This article describes how feature selection for learning to rank algorithms has become an interesting issue. While noisy and irrelevant features influence performance, and result in an overfitting problem in ranking systems, reducing the number of features by illuminating irrelevant and noisy features is a solution. Several studies have applied feature selection for learning to rank, which promote efficiency and effectiveness of ranking models. As the number of features and consequently the number of irrelevant and noisy features is increasing, systematic a review of Feature selection for learning to rank methods is required. In this article, a framework to examine research on feature selection for learning to rank (FSLR) is proposed. Under this framework, the authors review the most state-of-the-art methods and suggest several criteria to analyze them. FSLR offers a structured classification of current algorithms for future research to: a) properly select strategies from existing algorithms using certain criteria or b) to find ways to develop existing methodologies.

Athlete performance and injury prevention depend on computational biology to evaluate physiological responses and biomechanics to investigate movement. Incorporating these areas with deep learning improves risk evaluation, allowing for data-driven training tactics. Conventional injury risk evaluations are frequently incorrect, depending on subjective evaluations. Existing models do not capture intricate interactions between biological and biomechanical factors, requiring a more accurate, data-driven method. Objectives: This research proposes the AIRADL (Athlete Injury Risk Assessment using Deep Learning) Framework to forecast injury risk by incorporating physiological and biomechanical data, enhancing classification accuracy, and assisting training decisions. The proposed AIRADL model is trained on the Athlete Health & Motion Analysis Dataset (AHMAD), which contains physiological data such as heart rate, oxygen level, lactate level, and muscle fatigue, in addition to biomechanical factors such as stride length, joint flexibility, and movement symmetry. The data preprocessing steps include mean imputation for missing numerical values, mode imputation for categorical values, label encoding for categorical features, Min-Max scaling for normalization, and Chi-Square feature selection to maintain the most pertinent predictors. The dataset is split into 80% training and 20% testing. A DL4J MLP Classifier is employed to learn trends and classify performance risk levels. The AIRADL framework employs a three-hidden-layer DL4J MLP architecture consisting of 64–128–64 neurons with ReLU activation, Adam optimizer, and early stopping to prevent overfitting. The Athlete Health & Motion Analysis Dataset (AHMAD), a privately collected dataset, was used for experimentation, and evaluation metrics included accuracy, F1-score, and Matthews Correlation Coefficient (MCC), which measures the strength and balance of predictions. Experimental validation confirms model stability, demonstrating consistent convergence behaviour and superior performance compared to baseline methods. The AIRADL model attained 92.3% accuracy in high-risk classification, with precision, recall, and F1-score of 91.8%, 89.6%, and 90.7%, respectively. The MCC score was 89.2%, indicating excellent predictive ability, with lactate level, muscle fatigue, and movement symmetry being important risk indicators. Conclusion: AIRADL shows deep learning's capability in athlete risk prediction and provides a powerful tool for injury prevention.

Deep learning (also known as deep structured learning, hierarchical learning, or deep machine learning) is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and nonlinear transformations. Deep learning has been characterized as a class of machine learning algorithms with the following characteristics [257]: • They use a cascade of many layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The algorithms may be supervised or unsupervised and applications include pattern analysis (unsupervised) and classification (supervised). • They are based on the (unsupervised) learning of multiple levels of features or representations of the data. Higher-level features are derived from lower-level features to form a hierarchical representation. • They are part of the broader machine learning field of learning representations of data. • They learn multiple levels of representations that correspond to different levels of abstraction; the levels form a hierarchy of concepts. These definitions have in common: multiple layers of nonlinear processing units and the supervised or unsupervised learning of feature representations in each layer, with the layers forming a hierarchy from low-level to high-level features. Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks (DBN), and recurrent neural networks have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition, and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. In this chapter we start in Section 7.1 with an introduction, giving a brief history of this field, the relevant literature, and its applications. Then we study some basic concepts of deep learning such as convolutional neural networks, recurrent neural networks, backpropagation algorithm, restricted Boltzmann machines, and deep learning networks in Section 7.2. Then we illustrate three examples for Apache Spark implementation for mobile big data (MBD), user moving pattern extraction, and combination with nonparametric Bayesian learning, respectively, in Sections 7.3 through 7.5. Finally, we have summary in Section 7.6.

The rapid growth of the Internet of Things (IoT) and its extensive use in many regions, such as smart homes, healthcare, and vehicles, have made IoT security increasingly critical. Ransomware is an advanced and adjustable threat influencing users globally, limiting admittance to their data or systems over models like file encryption or screen locking. Traditional ransomware detection methods frequently drop, deprived of the ability to combat these threats successfully. Therefore, an effective and reliable mechanism is needed for ransomware detection. Deep learning (DL) and machine learning (ML) methods are very efficient and enhance model efficacy, offering burgeoning research paths, mainly in the ransomware detection realm, and presenting advantageous possibilities for new solutions. This study proposes a novel Multi-head Attention-Based Recurrent Neural Network with Enhanced Gorilla Troops Optimization for Cybersecurity Ransomware Detection (MHARNN-EGTOCRD) approach. The main goal of the MHARNN-EGTOCRD approach is to detect and classify ransomware attacks using advanced hybrid and optimization models in IoT environments. In the data normalization stage, the min-max normalization transforms input data into a suitable format. The dung beetle optimization (DBO) model is employed for the feature selection procedure to eliminate irrelevant, redundant, or noisy features. In addition, the proposed MHARNN-EGTOCRD model also implements a multi-head attention mechanism hybrid with a long short-term memory (MHA-LSTM) model for detecting ransomware. Finally, the hyperparameter selection of the MHA-LSTM model is performed by utilizing the EGTO model. The experimental analysis of the MHARNN-EGTOCRD technique is established on a ransomware detection dataset. The experimental validation of the MHARNN-EGTOCRD technique portrayed a superior accuracy value of 98.53% over existing models.

The American College of Cardiology (ACC)/American Heart Association (AHA) Task Force on Practice Guidelines was formed to make recommendations regarding the diagnosis and treatment of patients with known or suspected cardiovascular disease. Coronary artery disease (CAD) is the leading cause of death in the United States. Unstable angina (UA) and the closely related condition non–ST-segment elevation myocardial infarction (NSTEMI) are very common manifestations of this disease. These life-threatening disorders are a major cause of emergency medical care and hospitalizations in the United States. In 1996, the National Center for Health Statistics reported 1 433 000 hospitalizations for UA or NSTEMI. In recognition of the importance of the management of this common entity and of the rapid advances in the management of this condition, the need to revise guidelines published by the Agency for Health Care Policy and Research (AHCPR) and the National Heart, Lung and Blood Institute in 1994 was evident. This Task Force therefore formed the current committee to develop guidelines for the management of UA and NSTEMI. The present guidelines supersede the 1994 guidelines. The customary ACC/AHA classifications I, II, and III summarize both the evidence and expert opinion and provide final recommendations for both patient evaluation and therapy: Class I: Conditions for which there is evidence and/or general agreement that a given procedure or treatment is useful and effective . Class II: Conditions for which there is conflicting evidence and/or a divergence of opinion about the usefulness/efficacy of a procedure or treatment. Class IIa: Weight of evidence/opinion is in favor of usefulness/efficacy. Class IIb: Usefulness/efficacy is less well established by evidence/opinion. Class III: Conditions for which there is evidence and/or general agreement that the procedure/treatment is not useful/effective and in some cases may be harmful. The weight of the evidence was ranked highest (A) if the data …

Advances in data collection have resulted in an exponential growth of high-dimensional microarray datasets for binary classification in bioinformatics and medical diagnostics. These datasets generally possess many features but relatively few samples, resulting in challenges associated with the “curse of dimensionality”, such as feature redundancy and an elevated risk of overfitting. While traditional feature selection approaches, such as filter-based and wrapper-based approaches, can help to reduce dimensionality, they often struggle to capture feature interactions while adequately preserving model generalization. Therefore, this paper introduces the Adaptive Cluster-Guided Simple, Fast, and Efficient (ACG-SFE) feature selection, a hybrid approach designed to address the challenges of high-dimensional microarray data in binary classification. ACG-SFE enhances the Simple, Fast, and Efficient (SFE) evolutionary feature selection model by integrating hierarchical clustering to dynamically group correlated features based on the optimal number of clusters determined by the Silhouette index, Davies-Bouldin score, and the feature-to-observation ratio while adaptively selecting representative features within clusters using mutual information and adjusting the selection threshold through a progress factor. This hybrid filter-wrapper approach improves feature interactions, effectively minimizing redundancy and overfitting while enhancing classification performance. The proposed model is assessed against four state-of-the-art evolutionary feature selection models on 11 high-dimensional microarray datasets. Experimental results indicate that ACG-SFE effectively selects a small yet pertinent feature subset, minimizing redundancy while attaining enhanced classification accuracy and F-measure. Furthermore, its reduced RMSE between train and test accuracy substantiates its capability to reduce overfitting, outperforming comparative models. These findings establish ACG-SFE as an effective feature selection model for handling high-dimensional microarray data in binary classification, enhancing classification accuracy while selecting minimal relevant features to reduce unnecessary complexity and the risk of overfitting.

Recent developments in driving technology have led to the creation of advanced driver assistance systems and progress towards fully autonomous vehicles. Cars equipped with radar technology can simultaneously detect multiple vulnerable road users, assessing their distance, speed, and approach angle. For autonomous vehicles to be deemed safe for public roads, they must effectively identify and classify these users. This study employs time–frequency analysis and deep learning techniques to classify spectrograms derived from targets. The training and testing datasets were generated using frequency‐modulated continuous‐wave (FMCW) radar signals operating at 77 GHz. A five‐layer convolutional neural network (CNN) was trained for this purpose. We investigated how different time window types and durations affect the Short‐Time Fourier Transform calculation and the CNN classification accuracy for each scenario. As the length of the time window increases, frequency resolution improves, enabling better differentiation between closely spaced frequencies and enhancing classification accuracy. However, increased time window lengths lead to decreased time resolution, causing accuracy to plateau at 800; beyond this point, accuracy declines. We achieved an accuracy rate of 88.95% in classifying seven data classes, with improvements in specific classes compared to prior studies. The findings suggest that micro‐Doppler‐based convolutional neural networks can effectively classify vulnerable road users, contributing to collision avoidance efforts.

A comprehensive deep learning model for motor phenotypes of Parkinson's disease using three-dimensional kinect V2 detectors.

Android applications often contain vulnerabilities that can be exploited by malicious software, making automated malware detection is crucial. Since software vulnerabilities can have severe repercussions, a great deal of research and development effort has been put into developing mitigation strategies. Over the past few years, a number of different approaches have been tried in an effort to lessen the danger posed by vulnerabilities in software. The proposed approach involves multiple stages, including APK recompilation, feature extraction from manifest permissions, model training, testing, and performance analysis. The process begins with APK recompilation, where Android package files are decompiled using specialized tools to extract critical components such as manifest.xml, which contains permissions requested by the application. These permissions serve as key indicators of potential malicious intent. The extracted XML-based permissions and additional metadata are then transformed into structured feature vectors. Feature selection techniques are applied to retain the most relevant attributes, reducing noise and enhancing classification accuracy. The core classification module leverages Recurrent Neural Networks (RNNs) to analyze the extracted features and identify malicious patterns. The model undergoes supervised training using a labeled dataset comprising both benign and malicious APKs. During training, the RNN learns temporal dependencies and feature interactions, improving its ability to detect sophisticated malware. Once trained, the model is tested on a separate dataset to evaluate its classification performance. Various evaluation metrics such as accuracy, precision, recall, and F1-score are used to measure effectiveness. Finally the proposed model is compared with traditional machine learning classifiers and existing deep learning techniques. The results demonstrate that RNN-based classification significantly enhances malware detection accuracy due to its ability to process sequential patterns within application permissions. This research contributes to the field of Android security by presenting an efficient, scalable, and automated malware detection framework, leveraging deep learning for real-time classification of malicious applications.

This study employs deep learning techniques to present a compelling approach for modeling brain connectivity in EEG motor imagery classification through graph embedding. The compelling aspect of this study lies in its combination of graph embedding, deep learning, and different brain connectivity types, which not only enhances classification accuracy but also enriches the understanding of brain function. The approach yields high accuracy, providing valuable insights into brain connections and has potential applications in understanding neurological conditions. The proposed models consist of two distinct graph-based convolutional neural networks, each leveraging different types of brain connectivities to enhance classification performance and gain a deeper understanding of brain connections. The first model, Adjacency-based Convolutional Neural Network Model (Adj-CNNM), utilizes a graph representation based on structural brain connectivity to embed spatial information, distinguishing it from prior spatial filtering approaches dependent on subjects and tasks. Extensive tests on a benchmark dataset-IV-2a demonstrate that an accuracy of 72.77% is achieved by the Adj-CNNM, surpassing baseline and state-of-the-art methods. The second model, Phase Locking Value Convolutional Neural Network Model (PLV-CNNM), incorporates functional connectivity to overcome structural connectivity limitations and identifies connections between distinct brain regions. The PLV-CNNM achieves an overall accuracy of 75.10% across the 1-51 Hz frequency range. In the preferred 8-30 Hz frequency band, known for motor imagery data classification (including α, μ, and β waves), individual accuracies of 91.9%, 90.2%, and 85.8% are attained for α, μ, and β, respectively. Moreover, the model performs admirably with 84.3% accuracy when considering the entire 8-30 Hz band. Notably, the PLV-CNNM reveals robust connections between different brain regions during motor imagery tasks, including the frontal and central cortex and the central and parietal cortex. These findings provide valuable insights into brain connectivity patterns, enriching the comprehension of brain function. Additionally, the study offers a comprehensive comparative analysis of diverse brain connectivity modeling methods.

Recent advances in machine learning are transforming medical image analysis, particularly in cancer detection and classification. Techniques such as deep learning, especially convolutional neural networks (CNNs) and vision transformers (ViTs), are now enabling the precise analysis of complex histopathological images, automating detection, and enhancing classification accuracy across various cancer types. This study focuses on osteosarcoma (OS), the most common bone cancer in children and adolescents, which affects the long bones of the arms and legs. Early and accurate detection of OS is essential for improving patient outcomes and reducing mortality. However, the increasing prevalence of cancer and the demand for personalized treatments create challenges in achieving precise diagnoses and customized therapies. We propose a novel hybrid model that combines convolutional neural networks (CNN) and vision transformers (ViT) to improve diagnostic accuracy for OS using hematoxylin and eosin (H&E) stained histopathological images. The CNN model extracts local features, while the ViT captures global patterns from histopathological images. These features are combined and classified using a Multi-Layer Perceptron (MLP) into four categories: non-tumor (NT), non-viable tumor (NVT), viable tumor (VT), and non-viable ratio (NVR). Using the Cancer Imaging Archive (TCIA) dataset, the model achieved an accuracy of 99.08%, precision of 99.10%, recall of 99.28%, and an F1-score of 99.23%. This is the first successful four-class classification using this dataset, setting a new benchmark in OS research and offering promising potential for future diagnostic advancements.

Human metapneumovirus (hMPV) is a significant cause of respiratory illness, particularly in children, elderly individuals, and immunocompromised patients. Despite its clinical relevance, hMPV poses diagnostic challenges due to its symptom similarity with other respiratory illnesses, such as influenza and respiratory syncytial virus (RSV), and the lack of specialized detection systems. Traditional diagnostic methods are often inadequate for providing rapid and accurate results, particularly in low-resource settings. This study proposes a novel deep learning framework, referred to as hMPV-Net, which leverages Convolutional Neural Networks (CNNs) to facilitate the precise detection and classification of hMPV infections. The CNN model is designed to perform binary classification by differentiating between hMPV-positive and hMPV-negative cases. To address the lack of real-world patient data, simulated image datasets were used for model training and evaluation, allowing the model to generalize to various clinical scenarios. A key challenge in developing this model is the imbalance within the dataset, where hMPV-positive cases are often underrepresented. To mitigate this, the framework incorporates advanced techniques such as data augmentation, weighted loss functions, and dropout regularization, which help to balance the dataset, improve model robustness, and enhance classification accuracy. These techniques are crucial in addressing issues such as overfitting and generalization, which are common when working with limited datasets in medical imaging tasks. The dataset used for model training and testing consists of 10,000 samples, with an equal distribution of hMPV-positive and hMPV-negative cases. Experimental results demonstrate that the hMPV-Net model achieves a high test accuracy of 91.8%, along with impressive test precision, recall, and F1-score values around 92%. These metrics indicate that the model performs exceptionally well in classifying both hMPV-positive and hMPV-negative cases. Furthermore, the model exhibits superior computational efficiency, requiring only 3.2 GFLOPs, which is significantly lower than other state-of-the-art models such as ResNet-50 and VGG-16. This reduction in computational cost makes the model suitable for deployment in resource-constrained healthcare environments, where computing power and infrastructure may be limited.

A novel classification method for GPR B-scan images based on weak-shot learning