Data Classification Research Articles

Impact Analysis of Filter and Wrapper-Based Feature Selection Techniques for Webpages Phishing Attacks Identification

Phishing, which involves fraudulently gaining access to sensitive assets of unsuspecting individuals through deceptive and malicious emails, is a major global threat to internet users. The proliferation of phishing sites and their operations is occurring at an alarming rate, raising significant concerns about how to forestall them. Numerous research efforts are underway to detect phishing attempts before they can compromise important information and cause damage. Compared to conventional methods, machine learning has proven highly effective at detecting phishing attacks by analyzing different features. This study analyzed the behaviors of seven classification data mining algorithms on optimal subset features selection using Wrapper (Boruta) and Filter-based (Mutual-Information). Real-life phishing webpage datasets were used for the analysis. Ensemble classifiers such as Voting, Gradient Boosting, and Random Forest were used in the experiments. Two experiments were conducted. In the first experiment, K-Nearest Neighbor (K-NN) yielded the highest accuracy among single classifiers, with a score of 94.1%, while Random Forest (RF) ensemble achieved 96.7%. In the second experiment, using another baseline feature set, RF performed excellently under the Boruta method with an accuracy of 97.25%, while K-NN retained the highest accuracy of 95.20% among single classifiers. This study provides empirical evidence that feature selection techniques have a great impact on the performance of ML models, for both single and ensemble classifiers, in the detection of phishing attacks.

Comparative Analysis of Classification and Segmentation Performance of Different Dog Breeds Based on Mask R-CNN

The classification of different image data is a common task in real life, which has been achieved as early as convolutional neural networks. With the emergence of R-CNN (Region CNN), simple classification tasks began to develop towards object detection tasks, followed by various segmentation tasks. After Faster R-CNN [4], Mask R-CNN [3] achieved instance segmentation while detecting objects. This algorithm model needs to distinguish different target objects in the image panel and learn a large number of characteristics to represent the details of each target object. In dog data images, there are differences and similarities in the characteristics of dog species, such as fur color, texture, and body shape. So in this article, the author selected and created two different dog datasets, representing targets with significant differences and similarities in features. Compare and analyze the performance of Mask R-CNN object detection and instance segmentation using these two dog datasets.

Energy efficient and reliable data transmission in WBAN using multivariate gradient divergence African buffalo optimization

SummaryWireless body area network (WBAN), a popular radio communication technique, tracks patients' health conditions remotely using small, low‐power, portable sensors positioned on the body to sense vital signs and send data to a control node. To sustain long‐term health monitoring, WBANs require significant energy from the nodes. This research proposes an energy efficient and reliable data transmission using multivariate gradient divergence African buffalo optimization (EEMGDABO) to minimize the delay of critical node data transmission by identifying the optimal path to the control node. The protocol comprises two key processes: winsorized correlative segmented symbolic regression with reinforcement learning (WCS‐SRRL) and the walrus optimization algorithm (WaOA) for accurate priority‐based classification of health data (normal, abnormal, and critical). The second process involves transferring these priority‐based data with reduced energy consumption and delay while increasing throughput by determining the best path from the critical node to the destination (hospital/doctor) using EEMGDABO, which calculates node distance and energy fitness value. Implemented in MATLAB with a health monitoring WBAN dataset, the proposed protocol demonstrates 23% less energy consumption, 12.33% less delay, and 23.44% higher network lifetime compared to existing optimization approaches.

Behavioural Indonesian disaster data classification in social media using KNN, random forest, and RNN in machine learning

This research focuses on the classification of Indonesian disaster data extracted from social media using three distinct machine learning models: K-Nearest Neighbors (KNN), Random Forest, and Recurrent Neural Network (RNN). The study evaluates and compares the performance of these models based on accuracy metrics. The KNN model demonstrates competency with an 82% accuracy, showcasing its ability to handle data based on nearest neighbors. However, there are indications of potential limitations in handling complex patterns or high-dimensional datasets. In contrast, Random Forest achieves a notable 94% accuracy, highlighting its effectiveness in combining decision trees to enhance performance and mitigate overfitting. The RNN model exhibits the highest performance, achieving 96% accuracy, attributed to its proficiency in understanding temporal relationships and sequential patterns in data, making it a robust choice for sequential datasets such as text or time series. The findings underscore the importance of selecting models tailored to specific dataset characteristics and desired classification analyses. Furthermore, future research directions involve exploring adaptations or optimizations of these models for diverse disaster types and social media data, aiming to develop more sophisticated and responsive models for disaster-related analysis in the Indonesian context.

Investigation and Application of Various Algorithms used in Object Detection and Classification in Image Data

In this article, we compare and analyze the YOLO (You Only Look Once) method, widely employed for object detection in digital image processing, with the Haar feature-based cascade classifier method implemented using the OpenCV library. YOLO, a deep learning–based approach, excels in real-time object detection and recognition applications. In contrast, the Haar method utilizes a traditional approach to rapidly identify features. However, significant performance differences exist between the two methods. Experimental results and performance analyses demonstrate that YOLO provides high accuracy rates and real-time processing speeds in object detection tasks. The code implementations presented in this study will be valuable to researchers new to digital image processing. Additionally, YOLO has shown high performance on large and complex datasets by leveraging GPU capabilities. Experiments with various YOLO versions (e.g., YOLOv4, YOLOv5, YOLOv7) have established it as one of the most suitable options for real-time applications, particularly due to its low latency and high accuracy.

A privacy-preserving expert system for collaborative medical diagnosis across multiple institutions using federated learning

Expert system recommendation assists the healthcare system to develop in real-time monitoring and diagnosis of patient conditions over several healthcare institutions. Privacy concerns, however, present significant problems since patient data leaks can lead to big effects including financial losses for hospitals and invasions of personal privacy for people. To address these issues, the research introduces a privacy-preserving collaborative medical diagnosis (CMD) method on a federated learning (FL). FL maintains patient privacy and data localization by spreading only model parameters, therefore enabling training models on remote datasets. The combination of Partially Homomorphic Cryptosystem (PHC) and Residual Learning based Deep Belief Network (RDBN) ensures an accurate and safe classification of patient physiological data. Experimental results show that the proposed method is successful in maintaining the diagnostic accuracy over numerous healthcare institutions and protecting privacy. The results show that the RDBN and PHC computations requires around 1000 ms and 150 ms, respectively for classification and privacy; the data transmission from the user to server and from server to user is 5 MB and 4 MB, respectively. Finally with a 30% reduction in overhead, the proposed approach offers an average increase in classification accuracy of 10% over multiple datasets.

WIMOAD: Weighted Integration of Multi-Omics data for Alzheimer's Disease (AD) Diagnosis.

As the most common subtype of dementia, Alzheimer's disease (AD) is characterized by a progressive decline in cognitive functions, especially in memory, thinking, and reasoning ability. Early diagnosis and interventions enable the implementation of measures to reduce or slow further regression of the disease, preventing individuals from severe brain function decline. The current framework of AD diagnosis depends on A/T/(N) biomarkers detection from cerebrospinal fluid or brain imaging data, which is invasive and expensive during the data acquisition process. Moreover, the pathophysiological changes of AD accumulate in amino acids, metabolism, neuroinflammation, etc., resulting in heterogeneity in newly registered patients. Recently, next generation sequencing (NGS) technologies have found to be a non-invasive, efficient and less-costly alternative on AD screening. However, most of existing studies rely on single omics only. To address these concerns, we introduce WIMOAD, a weighted integration of multi-omics data for AD diagnosis. WIMOAD synergistically leverages specialized classifiers for patients' paired gene expression and methylation data for multi-stage classification. The resulting scores were then stacked with MLP-based meta-models for performance improvement. The prediction results of two distinct meta-models were integrated with optimized weights for the final decision-making of the model, providing higher performance than using single omics only. Remarkably, WIMOAD achieves significantly higher performance than using single omics alone in the classification tasks. The model's overall performance also outperformed most existing approaches, highlighting its ability to effectively discern intricate patterns in multi-omics data and their correlations with clinical diagnosis results. In addition, WIMOAD also stands out as a biologically interpretable model by leveraging the SHapley Additive exPlanations (SHAP) to elucidate the contributions of each gene from each omics to the model output. We believe WIMOAD is a very promising tool for accurate AD diagnosis and effective biomarker discovery across different progression stages, which eventually will have consequential impacts on early treatment intervention and personalized therapy design on AD.

AI-based fully automatic analysis of retinal vascular morphology in pediatric high myopia

PurposeTo investigate the changes in retinal vascular structures associated with various stages of myopia by designing automated software based on an artificial intelligence model.MethodsThe study involved 1324 pediatric participants from the National Children’s Medical Center in China, and 2366 high-quality retinal images and corresponding refractive parameters were obtained and analyzed. Spherical equivalent refraction (SER) degree was calculated. We proposed a data analysis model based on a combination of the Convolutional Neural Networks (CNN) model and the attention module to classify images, segment vascular structures, and measure vascular parameters, such as main angle (MA), branching angle (BA), bifurcation edge angle (BEA) and bifurcation edge coefficient (BEC). One-way ANOVA compared parameter measurements between the normal fundus, low myopia, moderate myopia, and high myopia groups.ResultsThe mean age was 9.85 ± 2.60 years, with an average SER of -1.49 ± 3.16D in the right eye and − 1.48 ± 3.13D in the left eye. There were 279 (12.38%) images in the normal group and 384 (16.23%) images in the high myopia group. Compared with normal fundus, the MA of fundus vessels in different myopic refractive groups was significantly reduced (P = 0.006, P = 0.004, P = 0.019, respectively), and the performance of the venous system was particularly obvious (P < 0.001). At the same time, the BEC decreased disproportionately (P < 0.001). Further analysis of fundus vascular parameters at different degrees of myopia showed that there were also significant differences in BA and branching coefficient (BC). The arterial BA value of the fundus vessel in the high myopia group was lower than that of other groups (P = 0.032, 95% confidence interval [CI], 0.22–4.86), while the venous BA values increased (P = 0.026). The BEC values of high myopia were higher than those of low and moderate myopia groups. When the loss function of our data classification model converged to 0.09, the model accuracy reached 94.19%.ConclusionThe progression of myopia is associated with a series of quantitative retinal vascular parameters, particularly the vascular angles. As the degree of myopia increases, the diversity of vascular characteristics represented by these parameters also increases.

Adaptive boundary-aware artificial immune recognition system for data classification

In this paper, we introduce AIBARS (Adaptive Immune Boundary-Aware Recognition System), a novel learning algorithm designed to improve upon the traditional Artificial Immune Recognition System (AIRS). AIBARS addresses key limitations of AIRS by integrating a boundary-aware stimulation measure, an antigen-specific stopping criterion, and a more directed mutation approach. These innovations enable AIBARS to effectively navigate complex decision boundaries and reduce time complexity, particularly in high-dimensional datasets. Experimental results show that AIBARS achieves a 5-10% improvement in classification accuracy compared to AIRS. Additionally, AIBARS reduces memory cells by 20-30%, leading to more efficient data representation and a 15-25% reduction in computational costs. The algorithm also reduces processing time by 10-20%, making it a more efficient approach overall. Evaluations were conducted using a variety of datasets, including public domain datasets from the UCI Machine Learning Repository and synthetic datasets. The results indicate that AIBARS offers significant performance and efficiency gains over AIRS, while substantially reducing computational expenses.

Monitoring of Land Cover Changes Between 1990 and 2014 over Cyprus using Multi-Temporal Landsat Imagery

Determining changes in forest resources and, consequently, land cover/land use is crucial for sustainable forest planning. This study aims to determine changes in land use classes, including Forest areas, Agricultural areas, Settlement areas, Other non-forest areas, and Water bodies in the study area located in Cyprus between 1990 and 2014. The study utilized digital management plans, a high-resolution base map, and Landsat satellite data for the relevant years. Necessary preprocessing steps were applied to prepare the satellite data for classification. Initially, unsupervised classification was conducted on the images to determine the number of distinguishable sub-information classes. Subsequently, supervised classification was performed using the maximum likelihood algorithm with the provision of training areas. The sub-information classes obtained from the supervised classification were consolidated into five predetermined main information classes. Following the classification process, the accuracy of the classification for each image was determined. Accordingly, the classification accuracy of the 1990 Landsat 5 TM satellite image was found to be 92% (Overall Classification Accuracy: 92%, Overall Kappa Statistics: 0.9000), while the classification accuracy of the 2014 Landsat 8 OLI satellite image was 89.20% (Overall Classification Accuracy: 89.20%, Overall Kappa Statistics: 0.8650). Subsequently, change analysis was conducted using the post-classification comparison method, and changes in land use within the study area over a twenty-four-year period were examined.

Implementasi Data Augmentation untuk Klasifikasi Sampah Organik dan Non Organik Menggunakan Inception-V3

The surge in global waste, particularly in Indonesia, with a total of 36.218 million tons per year, has become an urgent issue. Challenges in waste management are increasingly complex due to the lack of public understanding and awareness in classifying types of waste. One systemic approach to address waste classification issues involves the use of machine learning technology to categorize waste into two main types: organic and non-organic. The data used in this study comes from a Kaggle website dataset comprising 25,500 entries. This research employs a transfer learning approach with the Inception-V3 architecture and data augmentation implementation. Transfer learning is chosen for its proven performance in image data classification, while data augmentation is implemented to introduce diversity to the dataset. The research stages include business understanding, data preprocessing, data augmentation, data modelling, and evaluation. The results show that the use of transfer learning with the Inception-V3 approach and data augmentation implementation achieves an accuracy rate of 94%, which falls into the excellent category.

Semantic Mapping of Landscape Morphologies: Tuning ML/DL Classification Approaches for Airborne LiDAR Data

The interest in the enhancement of innovative solutions in the geospatial data classification domain from integrated aerial methods is rapidly growing. The transition from unstructured to structured information is essential to set up and arrange geodatabases and cognitive systems such as digital twins capable of monitoring territorial, urban, and general conditions of natural and/or anthropized space, predicting future developments, and considering risk prevention. This research is based on the study of classification methods and the consequent segmentation of low-altitude airborne LiDAR data in highly forested areas. In particular, the proposed approaches investigate integrating unsupervised classification methods and supervised Neural Network strategies, starting from unstructured point-based data formats. Furthermore, the research adopts Machine Learning classification methods for geo-morphological analyses derived from DTM datasets. This paper also discusses the results from a comparative perspective, suggesting possible generalization capabilities concerning the case study investigated.

Prediction and Diagnosis of Breast Cancer using Machine Learning Techniques

Introduction: One of the most common types of cancer and a significant contributor to the high death rates among women is breast cancer. It usually occurs in women. It is crucial to acquire a diagnosis early in order to kill cancer from becoming worse. Objective: The traditional diagnosing procedure takes more time. A fast and useful option can apply Machine Learning Technique (MLT) to identify illnesses. However new technology creates a variety of high-dimensional data kinds particularly when it comes to health or cancer data. Methods: Data classification techniques like Machine Learning are efficient. Particularly in the medical field where such techniques are often utilised to make decisions via diagnosis and analysis. Using Wisconsin Breast Cancer Dataset, the proposed research was carried out (WBCD). Some of these issues may be solved using the feature selection approach. Results: This research analyses the classification accuracy of different MLT: Logistic Regression, Support Vector Machine, and K-Nearest Neighbour. According to experiment results, SVM has the best accuracy of all algorithms, at 97.12%. Conclusion: The mentioned prediction models are based on several supervised MLT. Tenfold cross validation is applied. Additionally, author also proposed a Flow chart of breast Cancer using MLT.

The PS4-likelihood ratio calculator: flexible allocation of evidence weighting for case-control data in variant classification

BackgroundThe 2015 American College of Medical Genetics/Association of Molecular Pathology (ACMG/AMP) variant classification framework specifies that case-control observations can be scored as ‘strong’ evidence (PS4) towards pathogenicity.MethodsWe developed the PS4-likelihood...

Mining and evaluation of adverse event signals for capmatinib based on the FAERS database.

To conduct a comprehensive data analysis based on the FDA's Adverse Event Reporting System (FAERS) to mine possible adverse event (AE) signals of Capmatinib, providing valuable references for its clinical application. Capmatinib was the primary suspected drug in the search of FAERS database from the second quarter of 2020 to the fourth quarter of 2023. Data processing, screening, and classification were performed using methods such as the Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-item Gamma Poisson Shrinker (MGPS). A total of 1,991 AE reports directly related to Capmatinib were screened, identifying 269 Preferred Terms (PTs) involving 26 System Organ Classes (SOCs). Besides the AEs recorded in the drug label (such as edema, nausea, fatigue, and dyspnea), the study unearthed other high-risk AEs not listed in the label, including Renal and urinary disorders, Vocal cord paralysis, and Ear and labyrinth disorders. Among these, renal and urinary disorders, and ear and labyrinth disorders had a higher frequency and intensity of signals, suggesting that their mechanisms of occurrence could be a future research direction. This study uncovered new potential AEs of Capmatinib based on the FAERS database, providing reference for its safe clinical use. Special attention should be given to the occurrence of ear and labyrinth disorders and renal and urinary disorders, primarily presenting as pseudo-acute kidney injury, during treatment.

Tennis action recognition and evaluation with inertial measurement unit and SVM

Action recognition in tennis plays a crucial role for athletes and coaches, aiding in understanding and evaluating the players' skill levels to formulate more effective training plans and tactical strategies. To enhance the recognition and grading of tennis player actions, this study introduces the use of inertial measurement units and flexible resistive sensors for data collection. An improved Support Vector Machine is employed for data classification to achieve efficient action recognition. The results demonstrated that the proposed classification algorithm achieved an average accuracy of 95.35 % in recognizing actions of elite athletes, with the highest accuracy (96.38 %) observed in forehand strokes. In the case of sub-elite athletes, the algorithm achieved an impressive average accuracy of 97.67 %. For amateur enthusiasts, the algorithm exhibited an average accuracy of 94.08 %. Furthermore, elite athletes exhibited larger peak values in the three-axis acceleration waveform during ball striking. Specifically, the absolute peak value of acceleration in the Y-axis for elite athletes reached 78 m/s², representing an increase of 39 m/s² and 8 m/s² compared to the other two levels of athletes, respectively. Additionally, on the X and Z axes, elite athletes' acceleration peak values reached 59 m/s² and 78 m/s², significantly higher than those of sub-elite athletes and amateur enthusiasts. Moreover, the acceleration curves of elite athletes demonstrated a higher overall regularity. These findings indicate that the proposed action recognition method has a significant impact on recognition and evaluation, providing valuable insights for action recognition and assessment across various domains and advancing the application of artificial intelligence technology in the field of sports.

Classification of Simulated Fake Bandwidth Data Using LSTM

Hardly will someone acknowledge that the bandwidth we use every day is as authentic as most ISPs advertise, even those offering dedicated services. There are usually shortcomings, especially on upload and download bandwidth speeds. This paper presents the classification of simulated fake bandwidth data using the Long Short-Term Memory model, which though seldom found, is a very effective approach in network analysis. There were 1400 bandwidth data points collected from the MikroTik RB 1100 AHx device in a month, then further processed with normalization, and divided to have 80% training and 20% testing. The LSTM model applied had an accuracy rate of 98.93%, proving that it is capable of classifying either genuine or fake bandwidth instances accordingly. Of 1,400 test data points, the model managed to classify 723 as fake bandwidth and another 677 as genuine, resulting in a classification error rate of only 1.07%. The results clearly prove that LSTM has huge potential for real-time bandwidth manipulation detection, key to enhancing trust and efficiency in network management. In this respect, this research shows that bandwidth analysis combined with LSTM can be an original solution for network monitoring

NÂNG CAO HIỆU QUẢ PHÂN LỚP DỮ LIỆU KHÔNG CÂN BẰNG SỬ DỤNG KỸ THUẬT TĂNG MẪU THIỂU SỐ VÀ ĐẶC TRƯNG CỦA MỖI CỤM

This paper proposes a method to enhance the effectiveness of classifying imbalanced data. The main contribution of the method is integrating the K-means clustering algorithm and the minority oversampling technique VCIR to generate synthetic samples that closely represent the actual data characteristics. Experimental results have shown that the proposed method performs better on several metrics than current popular methods for handling imbalanced data, such as SMOTE, Borderline-SMOTE, Kmeans-SMOTE, and SVM-SMOTE. Keywords: Data classification; imbalanced data; oversampling; K-Means; SMOTE.

Classification of hyperspectral and LiDAR data by transformer-based enhancement

ABSTRACT The integration of multi-modal data allows for a more accurate representation of the ground characteristics. For a comprehensive interpretation of remote sensing data, existing multi-modal data fusion research mainly focuses on the joint utilization of 3D Light Detection and Ranging (LiDAR) and 2D Hyperspectral Image (HSI) data. However, existing algorithms do not pay much attention to the interaction of high-level semantic information between different modal data before fusion. This paper proposes a novel multi-modal data fusion deep learning network with the Cross-Modal Self-Attentive Feature Fusion Transformer (SAFFT). The framework employs a multi-head self-attention layer to fuse various attention information from multiple heads, effectively enhancing advanced feature information from different modalities for comprehensive integration. Experimental results on the Houston 2013 dataset demonstrate the effectiveness of the proposed method, which achieves an overall accuracy (OA) of 94.3757% in classifying 15 semantic classes.

Training of Teachers for Cadet Classes: Approaches and Successful Practices

The article offers an analysis of approaches and successful practices of teacher training for cadet classes. Various groups of methods were used in the course of the study, including the method of comparative analysis of normative and legal documents; analysis, systematization and generalization of the data obtained; complex analysis, classification and structuring of information and statistical data; modeling method; generalization of research results. As a result, system solutions for the training of teachers of cadet classes have been developed and theoretically substantiated. It should be noted that one of the conceptual ideas of teacher training for cadet classes is the idea of continuity and individualization of professional development of teachers in the field of education and socialization of students. Continuous professional development of teachers, including teachers of cadet classes, is carried out through the integration of formal and non-formal education, as well as through self-education and self-education. Teachers of cadet classes should have the competencies necessary for the full implementation of professional functions in the organization of education and socialization of students, since the teacher of the cadet class works not only with students, but also with cadet families, focusing on issues of social protection, education, formation of healthy public opinion and mood. The tasks of the educator of the training course include studying the mood, requests and needs of child-parent relations in cadet families and systematically informing the management and specialists of various services about the urgent tasks of working with cadet parents. The training of teachers of cadet classes is focused on the formation of a number of skills among educators: to analyze the academic performance and involvement of students in the educational process, to build effective communication, to organize activities to prevent negative manifestations and offenses.