Search

The mandibular condyle is a rounded bony projection with an upper biconvex, oval surface in the axial plane. Its form differs significantly among different individuals and age groups. This study aims to address the variability in mandibular condyle morphology, which can be indicative of Temporomandibular Joint disorders (TMD). Given the clinical importance of accurate condyle characterization, we developed a novel detection method leveraging deep learning and feature selection technologies. This method explicitly employs the YOLOv8 network to initially identify the region of interest (ROI) in digital panoramic images. Subsequently, the MobileViT system extracts detailed features from these regions. We introduced a modified Mountaineering Team-Based Optimization Algorithm to refine the feature selection process, which efficiently isolates the most relevant features from the extracted set. Our experimental design involved a robust dataset of 3000 digital panoramic images, classified into four distinct morphological types: round, pointed, angled, and flat. We assessed the performance of our developed method through various metrics, focusing on its ability to detect and describe the morphology of the condyle. The results demonstrate a high capability of the model, achieving an accuracy of 81.5% in binary classification and 83.5% in multi-classification scenarios. [ABSTRACT FROM AUTHOR]

In recent years, e-sports has experienced a rapid surge in popularity, attracting a vast and diverse audience. As this industry continues to evolve, understanding and predicting e-sport engagement becomes increasingly vital for stakeholders, including game developers, tournament organizers, sponsors, and marketers. Machine learning classification algorithms offer a powerful approach to analyse and forecast user engagement in e-sports, thereby enabling the industry to tailor experiences to individual preferences and behaviours. Thus, this study investigates the level of engagement classification technique of data mining using predictive modelling operations with four different classes, namely strongly agree, either agree or disagree, disagree, and strongly disagree. Machine learning algorithms, particularly classification models, have proven to be effective in analysing large and complex datasets related to e-sport engagement. This study applies statistical techniques to categorize users based on 59 attributes of 106 instances to predict the engagement levels. By training on historical user data, six classification algorithms from two groups, namely bayes and rules, have been used to identify patterns and trends that are indicative of different engagement levels, with the accuracy ranges from 76% to 92%. For feature selection, the result shows that participating in activities, enjoying exchanging ideas, and playing with like-minded gamers were the top three ranking dimensions contributing to the level of engagement. Machine learning classification algorithms have the potential to revolutionize how e-sport engagement is understood and optimized. By analysing diverse data points and leveraging advanced predictive techniques, machine learning algorithms enable stakeholders to tailor e-sport experiences to individual preferences and behaviours, ultimately enhancing user engagement and satisfaction. [ABSTRACT FROM AUTHOR]

A composite terminal sliding mode controller (CTSMC) for a kind of uncertain nonlinear system (UNS) is developed in this study. The primary aim of the design is to enhance the control performance of the CTSMC by learning its unknown parameters using a newly fuzzy neural network (FNN). Firstly, the stability and convergence of CTSMC for UNS with known parameters are demonstrated. Secondly, since some parameters of actual UNS are unmeasurable, a self-organizing feature selection fuzzy neural network (SOFSFNN) is intended to approach these unknown parts. Finally, the CTSMC using SOFSFNN is applied to UNS. The outcomes demonstrate that it has minimal tracking error, good robustness, and the ability to dynamically modify the network structure. • An improved composite terminal sliding mode controller is designed for a class of uncertain nonlinear systems. • Self-organizing feature selection fuzzy neural network is newly developed. • Results confirm that the proposed controller provides superior tracking performance against uncertainties. [ABSTRACT FROM AUTHOR]

In real-world scenarios, datasets often lack full supervision due to the high cost associated with acquiring decision labels. Completing datasets by filling in missing labels is essential for preserving the valuable feature information of individual samples. Furthermore, in the era of big data, datasets tend to exhibit high dimensionality, which adds complexity to subsequent data processing. In this study, a new semi-supervised feature selection technique is introduced. Firstly, a fully supervised dataset is created by utilizing a local density decision-labeling algorithm to fill in missing decision labels within the semi-supervised dataset. Next, a fuzzy dependency-based feature selection approach is presented to find and keep the most pertinent characteristics for the finished datasets. Finally, the effectiveness and reliability of our proposed method are validated through a series of rigorous experiments. [ABSTRACT FROM AUTHOR]

Label distribution learning(LDL) has been received widespread attention as an effective learning paradigm in the field of data mining. However, existing feature selection algorithms for LDL are performed under static conditions or semi-dynamic conditions, and the dynamic information of data is not considered sufficiently. To address these issues, a feature selection algorithm for LDL under dynamic conditions is proposed, which takes full use of dynamic information from data. Specially, a novel rough set model called F-double-fuzzy rough set is created to handle dynamic LDL data, which is extended from F-fuzzy rough set. Then, F-fuzzy-condition entropy is defined to fuse information together in F-double-fuzzy rough set, which is considered as a measure for feature selection. Thirdly, a dynamic feature selection algorithm for LDL is proposed with the F-fuzzy-condition entropy. Fourthly, a novel method of covering set is proposed to simulate dynamic LDL data, and its merits are explained intuitively. At last, experiments on 14 datasets with six widely used metrics demonstrate that the performance of the proposed algorithm has advantages over the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Cyberattack classification involves applying deep learning (DL) and machine learning (ML) models to categorize digital threats based on their features and behaviors. These models examine system logs, network traffic, or other associated data patterns to discriminate between standard activities and malicious actions. Efficient cyberattack classification is vital for on-time threat detection and response, permitting cybersecurity specialists to categorize and reduce potential risks to a system. Handling class-imbalanced data in cyberattack classification using DL is critical for achieving exact and robust models. In cybersecurity databases, instances of normal behavior frequently significantly outnumber instances of cyberattacks, foremost due to biased methods that may complete poorly on minority classes. To address this issue approaches such as oversampling the lesser class, undersampling the popular class, or using more advanced systems can be used. These plans defend that the DL technique is more complex when determining cyberattacks, so it increases complete performance and adapts the effect of the imbalance class on the classification results. This study presents a novel Hybrid Salp Swarm Algorithm with a DL Approach for Cyberattack Classification (HSSADL-CAC) technique. The HSSADL-CAC method intends to resolve class imbalance data handling with an optimum DL model for the recognition of cyberattacks. At first, the HSSADL-CAC method experiences data normalization as a pre-processing stage. The HSSADL-CAC technique uses the ADASYN approach to handle class imbalance problems. In addition, the HSSADL-CAC technique applies an HSSA-based feature selection approach. The HSSADL-CAC technique detects cyberattacks using a deep extreme learning machine (DELM) model. Finally, the hyperparameter tuning of the ELM model takes place by utilizing the beluga whale optimization (BWO) model. The performance analysis of the HSSADL-CAC technique employs a benchmark database. The comprehensive comparison research indicates the superior performance of the HSSADL-CAC technique in the cyberattack detection procedure. [ABSTRACT FROM AUTHOR]

With the fast development of the Internet of Things (IoTs), phishing attacks are transferring to this domain because of the number of IoT devices and private data they can handle. IoT phishing suggests the malicious practice of targeting IoT devices and their users to steal sensitive data, gain unauthorized access, or compromise the safety and functionality of these devices. Phishing attacks in the framework of IoT take several procedures, comprising SMS phishing, email phishing, or even physical attacks on the devices themselves. Deep learning (DL) approaches are executed to improve the prevention and detection of IoT phishing attacks. Consequently, this article introduces a Dwarf Mongoose Optimization with deep learning for Phishing Attack Detection (DMODL-PAD) method in an IoT platform. The objective of the DMODL-PAD method is to utilize feature selection (FS) with a hyperparameter tuning strategy for automated phishing attack detection in the IoT platform. The DMODL-PAD method involves the DMO model, unlike the FS method. Afterwards, a hybrid stacked autoencoder (HSAE) model can execute the phishing attack detection process. The DMODL-PAD method uses a jellyfish search optimizer (JSO) for the hyperparameter tuning process. The experimentation validation of the DMODL-PAD technique has been examined under a benchmark database. The extensive outcomes portrayed the superior detection results of the DMODL-PAD technique over other recent methods. These results ensured that the DMODL-PAD technique could effectually detect phishing attacks from the IoT environment. [ABSTRACT FROM AUTHOR]

Gastric cancer (GC) is the third leading cause of cancer death worldwide. In the field of medicine, machine learning is widely used in genetic data mining and the construction of diagnostic models. This study proposed an intelligent model DERFS-XGBoost for rapid and accurate diagnosis of GC based on gene expression data. Firstly, the data of GC were collected and preprocessed. Secondly, ANOVA, t-test and fold chang (FC) were used to select genes that had significant differentially expressed genes (DEGs), and random forest (RF) was used to calculate their importance, and then sequential forward selection (SFS) was used to obtain the optimal feature subset. Finally, XGBoost was used for classification after synthetic minority oversampling technique (SMOTE) balanced between tumor and normal samples. In order to objectively evaluate the results, the 10-fold cross-validation and 10 repeated experiments were used in the experiment, and the average value of the evaluation indexes was used to evaluate the classification effect. Based on the experiment, DERFS-XGBoost model accuracy rate was 97.6%, precision was 100%, the recall rate was 97.3%, F1 was 99%, and the area under the ROC receiver operating characteristic curve AUC was 98.7%. The DERFS-XGBoost model has new characteristics which are different from existing diagnostic models, and has achieved a high classification effect with a small number of genes in comparison tests, which provides a new method and basis for the diagnosis of GC. [ABSTRACT FROM AUTHOR]

Although rare overall, salivary gland carcinomas (SGCs) are among the most common oral and maxillofacial malignancies. The aim of this study was to develop a machine learning-based model to predict the survival of patients with SGC. Patients in whom SGC was confirmed by histological testing and who underwent primary extirpation at the authors' institution between 1963 and 2014 were identified. Demographic and clinicopathological data with complete follow-up information were collected for analysis. Feature selection methods were used to determine the correlation between prognosis-related factors and survival in the collected patient data. The collected clinicopathological data and multiple machine learning algorithms were used to develop a survival prediction model. Three machine learning algorithms were applied to construct the prediction models. The area under the receiver operating characteristic curve (AUC) and accuracy were used to measure model performance. The best classification performance was achieved with a LightGBM algorithm (AUC = 0.83, accuracy = 0.91). This model enabled prognostic prediction of patient survival. The model may be useful in developing personalized diagnostic and treatment strategies and formulating individualized follow-up plans, as well as assisting in the communication between doctors and patients, facilitating a better understanding of and compliance with treatment. [ABSTRACT FROM AUTHOR]

Ontology matching plays a pivotal role in aligning entities across diverse ontologies, enhancing interoperability and data exchange. Similarity features are the foundation of ontology matching, which assess entity similarities from multiple perspectives. Selecting relevant similarity features tailored to specific matching tasks is crucial due to the intricate heterogeneity of entities. However, traditional similarity feature selection methods face significant limitations, such as reliance on extensively curated datasets, the need for extensive human intervention, and inconsistencies in handling complex high-dimensional data structures. To address these challenges, we develop a novel transfer learning approach for automatic similarity feature selection. Our approach includes a new transfer learning framework to enhance similarity feature selection for ontology matching, a novel similarity feature alignment extraction method utilizing mutual information to align feature subsets using an entropy-based metric, and a real-value compact genetic algorithm to adjust the feature space and train an ordered weighting aggregation-based classifier. We compared our method with advanced techniques on OAEI's Benchmark and Bio-ML datasets. Experimental results show that our method significantly outperforms the competitors across various tasks. By automating similarity feature selection and reducing dependency on manual interventions, our method addresses the complexities of traditional approaches, enhancing the practical utility of ontology matching. [ABSTRACT FROM AUTHOR]

E-commerce platforms are recognizing the value of customer experience and are dedicating sections for customers to share reviews of the product purchased. Therefore, this study aimed to analyze Online Customer Review (OCR) to identify hidden emotion expressed about the purchasing experience and further identify factors relating to the product. Text-based emotion classification is a prominent and growing field to better understand human emotions. An integrated Information Gain-Recursive Feature Elimination (IG-RFE) and stacking ensemble learning were implemented to develop a predictive emotion classification model to identify the hidden emotions of the customers. Additionally, the Latent Dirichlet Allocation (LDA) model was used to extract the influencing factors, providing further insight in OCR. The study extracted eight emotions from OCR and seven influencing factors from product's attributes. The emotions included anger, anticipation, disgust, fear, happiness, sadness, surprise, and trust while the identified factors were quality, brand credibility, product functionality, usability, appearance, price, and functional effect. The extracted emotions and factors from the OCR provided valuable knowledge on the study. The findings showed knowledge gaps in emotion classification and customer behavior fields, suggesting further investigation for future study. [ABSTRACT FROM AUTHOR]

As autonomous vehicles grow more common, maintaining their cyber security becomes increasingly important. The CAN (Controller Area Network) bus, a critical communication network in self-driving cars, is susceptible to cyber-attacks that can jeopardize vehicle safety and performance. In this paper, we offer a novel hybrid approach, DeepXG, that combines XGBoost and deep learning (DL) approaches to detect intrusions in the CAN bus. Our model takes advantage of both algorithms' strengths to extract critical characteristics and learn complicated patterns for accurate and resilient intrusion detection. We conducted comprehensive studies to evaluate DeepXG's performance using a genuine CAN traffic dataset from a CAV's OBD-2 port. The proposed method outperformed many intrusion detection methods, achieving an amazing accuracy of 99.90%. The XGBoost feature relevance score enables effective feature selection while reducing computing complexity and boosting generalization. Our findings show that DeepXG helps improve cyber security in autonomous vehicles. The hybrid model's ability to effectively detect and classify network intrusions makes it a potential approach for safeguarding the CAN bus and ensuring autonomous vehicle safety. [ABSTRACT FROM AUTHOR]

Copyright of Afyon Kocatepe University Journal of Science & Engineering / Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi is the property of Afyon Kocatepe University, Faculty of Science & Literature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Internet Protocol version 6 (IPv6) is the most recent iteration of IP, designed to accommodate hundreds of thousands of devices with unique IP addresses. IPv6 introduces new features, such as the Neighbor Discovery Protocol (NDP) and Address Auto-configuration Scheme. For its effective operation, IPv6 relies on several protocols, including ICMPv6, which carries significant responsibilities. Similar to IPv4, IPv6 is susceptible to various attacks, including newer types like DDoS attacks executed via ICMPv6 messages, posing serious security and financial threats. Consequently, an Intrusion Detection System (IDS) is essential to safeguard against these attacks. IDS continuously evolve to incorporate features that can accurately detect such threats. However, feature selection strategies, particularly bio-inspired algorithms, often yield incorrect subsets of features. During machine learning processes, these inaccuracies impede the detection accuracy of DDoS attacks using ICMPv6. Many Optimization Search Algorithms become trapped in local minima and fail to consider multi-objective approaches, resulting in suboptimal feature selection. To address this, optimizing a bio-inspired algorithm within an IPv6 network has been proposed. Specifically, hybridizing the MFPA algorithm with the PSO algorithm is suggested to enhance detection accuracy. The selected features are used to train the dataset with a Support Vector Machine (SVM) classifier. The proposed approach is evaluated using the ICMPv6 dataset on various attacks, demonstrating superior classification accuracy of 97.99 %. It also reduced the number of features from 19 to 8, showcasing its efficiency. [ABSTRACT FROM AUTHOR]

Bacterial pharyngitis and tonsillitis can lead to severe complications if untreated, while oral cancer poses risks of spreading if not detected early. Classifying pharyngitis, oral cancer, and tonsillitis is essential for timely and accurate medical interventions. Early diagnosis enables effective treatment planning, potentially saving lives and minimizing complications. Classification aids in tailoring specific medical approaches, contributing to better patient outcomes and healthcare management. The global COVID-19 pandemic has prompted a renewed interest in telemedicine for managing these conditions. Existing techniques, like Bag of Visual Words and individual pre-trained models, fall short in achieving optimal classification. To address this, we propose GHBResIncep, a framework that combines deep features from SE-ResNet, Inception-v4 and EfficientNetV2 with shallow features from improved BOVW, HOG, and Gabor filters. Employing an iterative relief feature weighting algorithm enhances feature selection. Our proposed model achieved highest classification accuracy of 95.58 %.The multi-fused features are fed into ML classifiers like XGBoost, showcasing significant improvements in various classification parameters compared to prior works. [ABSTRACT FROM AUTHOR]

High-precision load forecasting serves as the foundation for power grid scheduling planning and safe economic operation. In scenarios where only historical power load data is available without other external information, fully exploiting meaningful features from the temporal load sequence is crucial for improving the accuracy of load forecasting. Therefore, an ultra-short-term load forecasting method that combines eXtreme gradient boosting (XGBoost) and bidirectional gated recurrent unit (BiGRU) is proposed in this paper. Considering various factors that affect loads, a candidate feature set is established, which includes temporal information and historical loads. XGBoost is used to select the features that contribute significantly to load forecasting, forming an optimal feature set. These optimal features are then used as inputs to the BiGRU, and the bayesian optimization algorithm is applied to optimize the network hyperparameters. Then the load forecasting model for the next 15 minutes based on BiGRU is generated by training iteratively. The proposed XGBoost-BiGRU method is validated on real load data from a province in China. Experimental results demonstrate that the method can effectively avoid the impact of redundant features, improving both prediction accuracy and efficiency. The research has significant importance for guiding real-time supply-demand balance calculations and scheduling in power grids. [ABSTRACT FROM AUTHOR]

The most common reason behind death all over the world is heart diseases. These conditions are to hit hardest in low- and middle-income nations, where 80% of premature heart attacks could be prevented. In this regard, early diagnosis also plays an important role in increasing patient health and survival rate from heart disease. The purpose of this study was to improve the forecasting power by means of feature selection techniques and then apply K-Fold cross validation in combination with high-performance ensemble machine learning (ML) methods (J48, Artificial Neural Networks (ANNs), Logistic Regression, Naive Bayes, K-Nearest Neighbors) by utilizing a dataset of 401,958 patients. Our experimental results demonstrate that ANNs achieve the highest accuracy at 91.48%. They also record the lowest Mean Absolute Error (MAE) of 0.13, highlighting their precision in predictions. Additionally, ANNs exhibit a low root Mean Squared Error (RMSE) of 0.26, further indicating their reliability in modeling. [ABSTRACT FROM AUTHOR]

Metamaterial absorbers (MMAs) have tremendous potential for controlling and modulating Terahertz electromagnetic (EM) waves. It is challenging to design MMAs for optimal performance using conventional methods due to their time-consuming and computationally demanding nature. Machine learning (ML) regressor approaches have emerged as potential tools for optimizing the predictive modeling of MMAs in recent years. This article examined different regression techniques, such as Decision Tree, k-Nearest Neighbors, Random Forest, Extra Trees (ET), Extreme Gradient Boosting, Bagging, and Categorical Boosting. The primary goal is to assess the effectiveness of each regressor technique in forecasting the performance of MMAs using various performance metrics. The study focuses on a representative spectrum of MMAs, chosen for its key features that are common across a wide range of MMAs. The extra tree Regressor outperforms other models in the comparative analysis due to its capacity to integrate multiple decision trees and include randomness into the feature selection process, resulting in improved predictive skills. The ET regressor forecasts MMA performance remarkably well for a 70 % test size, with a Root Mean Squared Error of 0.0251, a Mean Absolute Error of 0.0074, a Mean Squared Error of 0.006, and an Adjusted R-squared of 0.9873. These findings can assist researchers in optimizing metamaterial absorber designs for Terahertz applications using ML regressor techniques and provide insights into how these methods can be generalized to other spectra with similar characteristics. [ABSTRACT FROM AUTHOR]

Copyright of Informatica (03505596) is the property of Slovene Society Informatika and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

WSNs are integral to various applications, ranging from environmental monitoring to industrial automation. However, their vulnerability to malicious activities necessitates robust security measures. The proposed Ensemble Intrusion Detection System (ENS-IDS) leverages machine learning techniques to detect anomalies in the WSN data, identifying potential intrusions or security breaches. The system incorporates feature selection, model training, and real-time monitoring to enhance its accuracy and responsiveness. Evaluation metrics, including precision, recall, and F1 score, demonstrate the effectiveness of the ENS-IDS in mitigating security threats within the WSN environment. The presented ENS-IDS is evaluated on KDD and CICIDS2017 dataset and comparison on known classifiers such as SVM, random forest, extra tree, KNN, logistic regression, decision tree and ensemble classifiers such as XGBoost, CatBoost and LGBM. Our model ENS-IDS has given better accuracy, precision, recall and F1-score. [ABSTRACT FROM AUTHOR]

Breast cancer is still a dangerous and common disease that affects women all over the world, which highlights how crucial early identification is to better patient outcomes. In recent years, utilizing machine learning (ML) algorithms has improved accuracy and efficiency dramatically in a variety of applications, showing promising outcomes. This article provides a novel machine-learning approach to increase the accuracy of breast cancer detection. To improve diagnostic efficiency and accuracy, our suggested methodology combines sophisticated feature selection strategies, reliable classification algorithms, and enhanced model training methodologies. We investigated several ML classifiers, and after thorough hyperparameter tuning, the models were. Random forest and gradient boosting have achieved the highest performance with an accuracy of 97.90% and an ROC score of 0.99. This research highlights the effectiveness of ML, particularly the random forest algorithm, in breast cancer diagnosis and prognosis. Future work may explore deep learning techniques for determining the disorder's severity. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) sensor networks are connected systems of physical devices set with actuators, sensors, and communication abilities, allowing them to gather, spread, and exchange information with centralized methods. These networks are essential in numerous businesses, such as healthcare, manufacturing, agriculture, and smart cities, as they deliver real-time observation, data-driven insights, and automation. Threat recognition in IoT sensor networks is a vital feature of safeguarding the protection and consistency of interconnected systems in the IoT. As IoT sensor networks endure to increase across various industries, the vulnerability to malicious actions and cyber-attacks increases. Threat recognition utilizing deep learning (DL) leverages neural networks to examine complex patterns and anomalies in data, permitting the identification of potential safety threats. DL techniques like convolutional neural networks (CNNs) or recurrent neural networks (RNNs) excel at learning complex representations of data and feature extraction, making them suitable for identifying sophisticated attacks in different fields, including cybersecurity. This research develops a Hybrid Sine-Cosine Chimp Optimization Feature Selection with a Deep Learning (HSCCOFS-DL) approach for Threat Recognition in IoT Sensor Networks. The foremost aim of the HSCCOFS-DL system lies in the automated detection of threats using DL models. To accomplish this, the HSCCOFS-DL approach undergoes a data normalization process. Besides, the selection of features can be performed using the HSCCO algorithm. Meanwhile, the symmetrical autoencoder (SAE) technique effectively classifies threats. Finally, the sparrow search algorithm (SSA) can be applied to the selection of the hyperparameter of the SAE system. The experimental assessment of the HSCCOFS-DL technique takes place on a benchmark dataset. The simulation results indicated that the HSCCOFS-DL approach attains enhanced performance over other methods. [ABSTRACT FROM AUTHOR]

In the era of big data, there exist complex structure between different classes labels. Hierarchical structure, among others, has become a representative one, which is mathematically depicted as a tree-like structure or directed acyclic graph. Most studies in the literature focus on static feature selection in hierarchical information system. In this study, in order to solve the incremental feature selection problem of hierarchical classification in a dynamic environment, we develop two incremental algorithms for this purpose (IHFSGR-1 and IHFSGR-2 for short). As a preliminary step, we propose a new uncertainty measure to quantify the amount of information contained in the hierarchical classification system, and based on this, we develop a non-incremental hierarchical feature selection algorithm. Next, we investigate the updating mechanism of this uncertainty measure upon the arrival of samples, and propose two strategies for adding and deleting features, leading to the development of two incremental algorithms. Finally, we conduct some comparative experiments with several non-incremental algorithms. The experimental results suggest that compared with several non-incremental algorithms, our incremental algorithms can achieve better performance in terms of the classification accuracy and two hierarchical evaluation metrics, and can significantly accelerate the fuzzy rough set-based hierarchical feature selection. [ABSTRACT FROM AUTHOR]

Breast cancer, a prevalent global concern affecting women, underscores the importance of early detection for improved treatment outcomes and reduced mortality rates. Mammogram image is widely employed as a tool for early detection of breast tumors. Incorrect diagnoses elevate the risk of cancer metastasis to vital organs like the lungs, stomach and lymph nodes. This study presents a software application categorizing mammogram images as benign or malignant. It relies on intrinsic features and employs twelve pre-trained deep-learning models. Additionally, ten feature selection algorithms are utilized to identify crucial attributes. Exploiting various feature selection techniques, pinpoint the most representative ones. The selected features from each algorithm contribute to building and testing the Gaussian Support Vector Machine (SVM) classifier. ReliefF selects the optimal features, reflecting the highest test accuracy in the SVM classifier. The recorded results demonstrate an accuracy, sensitivity, precision and specificity of 99.9%. These findings underscore the potential of combining diverse deeplearning structures with feature-reduction techniques to enhance diagnostic capabilities. The research highlights the technology’s potential adoption in the healthcare sector, particularly considering the substantial volume of images involved and the heightened reliability it introduces to the mammogram image diagnosis process. [ABSTRACT FROM AUTHOR]

Wireless sensor network (WSN) consists of large number of sensor nodes that are deployed in geographical locations to collect sensed information, process data and communicate it to the control station for further processing. Due the unfriendly environment where the sensors are deployed, there exist many possibilities of malicious nodes which performs malicious activities in the network. Therefore, the security threats affect performance and life time of sensor networks, whereas various security aspects are there to address security issues in WSN namely Cryptography, Trust Management, Intrusion Detection System (IDS) and Intrusion Prevention Systems (IPS). However, IDS detect the malicious activities and produce an alarm. These malicious activities exploit vulnerabilities in the network layer and affect all layers in the network. Existing feature selection methods such as filter-based methods are not considering the redundancy of the selected features and wrapper method has high risk of overfitting the classification of intrusion. Due to overfitting, the classification algorithm fails to detect the intrusion in better manner. The main objective of this paper is to provide the efficient feature selection algorithm which was suitable for any type classification algorithm to detect the intrusion in an effective manner. This paper, the security of the network is addressed by proposing Feature Selection Algorithm using Chi Squared with Ensemble Method (FSChE). The proposed scheme employs the combination of decision tree along with the random forest classification algorithm to form ensemble classifier. The experimental results justify the feasibility of the proposed scheme in terms of attack detection, packet delivery ratio and time analysis by employing NSL KDD cup data Set. The obtained results shows that the proposed ensemble method increases the overall performance by 10% to 25% with respect to mentioned parameters. [ABSTRACT FROM AUTHOR]

With the fast growth of mobile phone usage, malicious threats against Android mobile devices are enhanced. The Android system utilizes a wide range of sensitive apps like banking apps; thus, it develops the aim of malware that uses the vulnerability of safety measures. Identifying Android malware in smartphones is a vital target for the cyber community to eliminate menacing malware instances. Drawing stimulus from the adaptability and efficacy of biological systems, these methods emulate nature's problem-solving systems for identifying malicious software. By integrating principles, namely, swarm intelligence (SI), neural networks (NN), and genetic algorithms (GA), these bioinspired systems reveal exceptional efficiency in identifying both known and developing Android malware attacks. This bioinspired system provides a capable avenue for robust Android malware detection from an ever-shifting threat landscape. This article designs a Bioinspired Artificial Intelligence-based Android Malware Detection and Classification (BAI-AMDC) technique for Cybersecurity Applications. The BAI-AMDC technique exploits the concept of bioinspired algorithms with a DL approach for the classification and detection of Android malware. In the BAI-AMDC technique, an improved cockroach swarm optimization algorithm-based feature selection (ICSOA-FS) technique can be applied to choose optimum features. The BAI-AMDC technique employs a bidirectional gated recurrent unit (BiGRU) model for Android malware detection. An arithmetic optimization algorithm (AOA) can be utilized to enhance the detection performance of the BAI-AMDC technique. The experimental validation of the BAI-AMDC system can be performed on the CICAndMal2017 database with 10,000 instances. The simulation values highlighted the productive ability of the BAI-AMDC system on the Android malware recognition process. [ABSTRACT FROM AUTHOR]

Chronic kidney disease (CKD) emerges as a global health problem with high morbidity and mortality rates, and it induces other diseases. Patients often fail to notice these diseases because there are no obvious symptoms during the earlier stage of CKD. Earlier diagnosis of CKD allows them to receive early medical intervention to improve the disease progression and outcome. Earlier stratification of CKD can affect the health care provided to patients through various options, such as kidney transportation, hemodialysis, or pharmacological care in milder cases. Recently, deep learning (DL) and machine learning (ML) approaches have gained significance in the domain of medical diagnoses due to their high prediction accuracy. DL and ML approaches can successfully aid physicians in achieving these goals due to their accurate and fast detection performance. The performance of the proposed technique mainly depends on selecting the suitable algorithms and appropriate features. Thus, the study introduces an Eurygasters Optimization Algorithm with an Ensemble DL CKD detection (EOAEDL-CKDD) method. The EOAEDL-CKDD method aims to detect and classify the presence of CKD using feature selection and hyperparameter tuning strategies. The EOAEDL-CKDD method applies a min-max scalar to convert the data into a uniform format to achieve this. The EOAEDL-CKDD technique exploits EOA to select features. This is followed by an ensemble of long short-term memory (LSTM), bidirectional gated recurrent unit (BiGRU), and bidirectional LSTM (BiLSTM) models that are used for the CKD detection process. Finally, a shuffled frog leap algorithm (SFLA) based hyperparameter selection process is carried out to choose the ensemble models' hyperparameter values optimally. The empirical assessment of the EOAEDL-CKDD method is tested on the benchmark CKD dataset. The experimental values highlighted that the EOAEDL-CKDD technique gains an optimal detection rate compared to existing models. [ABSTRACT FROM AUTHOR]

With so many thyroid knobs (nodules) discovered by accident, it is critical to recognize as many aberrant knobs (nodules) as possible from fine-needle aspiration (FNA) biopsies or other medical procedures while excluding those that are virtually certainly benign. Thyroid ultrasonography, on the other hand, is prone to interobserver variability and subjective translations. An effective deep learning model for segmenting and categorizing thyroid nodules in this study follows the stages below: data collection from a well-known archive, The Thyroid Digital Image Database (TDID), which comprises ultrasound pictures from 298 patients, preprocessing using anisotropic diffusion filter (ADF) for removing noise and enhancing the images, segmentation using a bilateral filter for segmenting images, feature extraction using grey level occurrence matrix (GLCM), feature selection using Multi-objective Particle Swarm with Random Forest Optimization (MbPSRA) and finally classification happens were Residual U-Net will be used. Experiment evaluation states the proposed model outperforms well than other state-of-art models. [ABSTRACT FROM AUTHOR]

Accurate forecasting is crucial for successfully integrating photovoltaic (PV) power plants into electrical grids and microgrids. Accordingly, this work presents a hybrid methodology for day-ahead PV power forecasting (PPF). It begins by examining three deep learning (DL) techniques, long short-term memory (LSTM), gated recurrent unit (GRU), and multilayer perceptron (MLP), as potential forecasting models. To find a robust forecasting model, feature selection is employed to select the most relevant input features, and additionally, hyperparameter optimization is performed using the Chu–Beasley genetic algorithm to automatically set the hyperparameters for each technique. An initial day-ahead PPF is computed recursively after selecting the optimal forecasting model. Subsequently, this initial forecast is refined using a long-short-term memory autoencoder (LSTM-AE) that corrects the initial PPF. To further enhance the interpretability of the final forecast, the k-means algorithm, incorporating a soft-dynamic time warping (DTW) metric, is utilized. The efficacy of the methodology is validated using real data from a solar farm at the State University of Campinas (UNICAMP) in Brazil. Empirical results demonstrate that the proposed methodology improves the forecast accuracy by more than 3.5% when LSTM-AE is applied for correction compared to state-of-the-art models. [ABSTRACT FROM AUTHOR]

The establishment of the NOx concentration prediction model is a prerequisite for the selective catalytic reduction (SCR) flue gas denitriflcation systems in thermal power plants to overcome measurement delays in continuous emission monitoring systems and achieve precise control. This paper proposes a NOx emission prediction method for thermal power plants based on the attention mechanism and the gated recurrent unit (Attention-GRU) to address the issues of low accuracy and cumbersome feature selection process in NOx emission prediction models established for the SCR system under complex working conditions. Firstly, this article utilizes the GRU to extract data features of NOx emissions from thermal power plants at the time scale, in order to establish a long-term prediction model. Secondly, the feature dimension clustering the k-means algorithm can effectively improve the learning efficiency of temporal features and reduce the computational complexity of subsequent algorithms. Finally, the parameter attention mechanism is introduced to autonomously select favorable temporal features for predicting NOx emissions in thermal power plants, replacing the complex and time-consuming data screening process. This paper has also collected historical datasets of the SCR system form a thermal power plant under complex and simple working conditions to verify the effectiveness of Attention-GRU. The experimental results show that compared with existing methods (see, e.g., ELM, RF, SVM, BPNN, LSTM, GRU, BiLSTM, and CNN-LSTM), our method has higher prediction accuracy of NOx emissions from thermal power plants, which helps to improve the control performance of SCR systems to reduce atmospheric pollution. At the same time, this article has also found that the attention mechanism can significantly reduce the dependence of existing methods on the feature selection process, effectively solve the interference problem of noise data on model feature extraction, and further improve the prediction accuracy of NOx emissions in thermal power plants. [ABSTRACT FROM AUTHOR]

Anomaly intrusion detection is a critical component of modern cybersecurity systems, aiming to identify and flag abnormal activities or behaviors that deviate from expected patterns within computer networks. Unlike signature-based intrusion detection systems that rely on known attack patterns, anomaly detection techniques focus on detecting unknown or novel attacks that lack predefined signatures. In recent years, machine learning and deep learning techniques have emerged as promising solutions to provide an additional layer of defense against emerging threats and zero-day attacks. This survey article provides a comprehensive review of the state of the art in network intrusion detection using ML and DL. We start by presenting an overview of the challenges and requirements associated with intrusion detection in today's dynamic network environments. We then delve into the fundamental concepts and methodologies of ML and DL, highlighting their strengths and limitations when applied to intrusion detection. We discuss the various types of network intrusion detection datasets commonly used in research, along with the preprocessing techniques employed to ensure data quality. We explore different feature selection and extraction methods that enable the effective representation of network traffic data, facilitating accurate intrusion detection. We review their architectural designs, training processes, and optimization techniques while discussing their performance in terms of detection accuracy. We highlight the current research trends and challenges in the field, including adversarial attacks, interpretability, scalability, and real-time processing. We conclude with potential future directions and recommendations for researchers and practitioners. [ABSTRACT FROM AUTHOR]

Diabetes, a pervasive global health concern, presents diagnostic challenges due to its nuanced onset and far-reaching implications. Traditional diagnostic approaches, reliant on time-consuming assessments, necessitate a paradigm shift towards more efficient methodologies. In response, this study introduces a diagnostic support system leveraging the power of optimized machine learning algorithms. Addressing class imbalance within a dataset comprising 768 records, our methodology intricately weaves together feature selection, dimensionality reduction techniques, and grid search optimization. Specifically, the Extra Trees model, fine-tuned via grid search, emerges as the most potent, showcasing remarkable performance metrics: an accuracy score of 92.5%, an F1-score of 93.7%, and an AUC-ROC of 92.47%. These findings underscore the pivotal role of machine learning in reshaping diabetes diagnosis, offering transformative possibilities for global healthcare enhancement. [ABSTRACT FROM AUTHOR]

Distributed Denial of Service (DDoS) attacks pose a significant threat to online services by flooding targets with unusually high volumes of traffic or data, disrupting services and causing financial and reputational damage. This paper explores the workings and impact of DDoS attacks, with a variety of methods used by attackers to exploit vulnerabilities in the target infrastructure. To address these risks, this paper advocates the application of Machine Learning (ML) techniques. ML allows computers to learn patterns from data and make decisions autonomously and offers a proactive defense against DDoS attacks. Using UNSW-NB15 Dataset, this study evaluates the performance of eight ML algorithms Logistic Regression, KNN, SVM, Random Forest, Decision Tree, LSTM, MLP, and GRU, the evaluation of this study is based on Accuracy, Precision, Recall, and F1 Score metrics. The results show that Random Forest is the most effective algorithm, showing superior performance across all metrics with 97.68% accuracy. This explains the potential of Machine Learning, specifically the Random Forest algorithm. can be considered to improve security and reduce the risk posed by DDoS attacks. [ABSTRACT FROM AUTHOR]

In the field of neural language processing, current research projects that apply meta-heuristic approaches to Arabic text are extremely limited given the complexity of this language in terms of structure, grammatical rules, morphology, syntactic analysis and derivation rules. In this paper, we propose a new approach for processing Arabic documents using the chaotic grey wolf optimization technique, which is employed as a supervised learning method by simulating the behavior of grey wolves in searching, circling and pursuing their prey. The proposed method is divided into four phases. The term weighting phase uses TF- IDFC-RF, which is based on TF-IDF and the relevance frequency of terms in documents and classes. The feature selection phase combines the chaotic method of the grey wolf optimization algorithm to both reduce and identify relevant features for model building. In the third phase, we use three classifiers: firefly algorithm, SVM and NB to classify documents into several classes. In the experimental process, we used two collections of reference documents to present a comparative study of the different term weighting systems. The experimental results show that the new architecture, based on the chaotic optimization of the grey wolf algorithm with the Firefly algorithm, obtained the best results in terms of accuracy. [ABSTRACT FROM AUTHOR]

Software-defined networking (SDN) is an approach to network management allowing to enhance the performance of the network and making it more flexible. The centralized architecture of SDN makes it vulnerable to cyberattacks, especially distributed denial of service (DDoS) attacks. Existing research investigates the detection of DDoS attacks separately on the control plane and data plane. However, there is a need for efficient and accurate detection of these attacks using features obtained from both control and data planes. Therefore, we present a mechanism for identifying DDoS attacks using entropy, multiple feature selection mechanisms, and deep learning. Initially, we use entropy on the control plane to detect anomalous activity and identify suspicious switches. Next, we capture traffic on the suspicious switches to detect DDoS attacks. To detect these attacks, we utilize multi-layer perceptron (MLP) deep learning models, convolutional neural network (CNN), and the long short-term memory (LSTM) approach. An InSDN dataset is used to train the model and test data are generated using Mininet emulation and the Ryu controller. The results reveal that LSTM outperforms MLP and CNN, achieving an accuracy of 99.83%. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a neurodegenerative disorder that influence brain's neurological, behavioral, and physiological functions and includes motor and nonmotor manifestations. Although there have been several PD diagnosis systems with supervised machine learning techniques, there are more efforts that need to enhance the accurate detection of PD in its early stage. The current paper developed a novel approach by integrating Least Squares Support Vector Regression (LS-SVR) and Fuzzy Clustering for Unified Parkinson's Disease Rating Scale (UPDRS) diagnosis. This paper used feature selection and Principal Component Analysis (PCA) to overcome the multicollinearity issues in data. This paper used a large medical dataset including Motor- and Total-UPDRS to demonstrate how the proposed method can improve prediction performance via extensive evaluations and comparisons with existing methods. Compared to other prediction methods, the experimental results demonstrate that the proposed method provided the best accuracy for Total-UPDRS (Root Mean Squared Error = 0.7348; R 2 = 0.9169) and Motor-UPDRS (Root Mean Squared Error = 0.8321; R 2 = 0.8756) predictions. [ABSTRACT FROM AUTHOR]

Attention Deficit Hyperactivity Disorder (ADHD) is a neurological condition, typically manifesting in childhood. Behavioral studies are used to treat the illness, but there is no conclusive way to diagnose it. To comprehend changes in the brain, electroencephalography (EEG) signals of ADHD patients are frequently examined. In the proposed study, we introduce EEG feature map (EEG-FM)-based image construction to input deep learning architectures for classifying ADHD. To demonstrate the effectiveness of the proposed method, EEG data of 15 ADHD patients and 18 control subjects are analyzed and detection performance is presented. EEG-FM-based images are obtained using both traditional time domain features used in EEG analysis, such as Hjorth parameters (activity, mobility, complexity), skewness, kurtosis, and peak-to-peak, and nonlinear features such as the largest Lyapunov Exponent, correlation dimension, Hurst exponent, Katz fractal dimension, Higuchi fractal dimension, and approximation entropy. EEG-FM-based images are used to train DarkNet19 architecture and deep features are extracted for each image dataset. Fewer deep features are chosen for each image dataset using the Minimum Redundancy Maximum Relevance (mRMR) feature selection method, and the concatenated deep feature set is created by merging the selected features. Finally, various machine learning methods are used to classify the concatenated deep features. Our EEG-FM and DarkNet19-based approach yields classification accuracies for ADHD between 96.6% and 99.9%. Experimental results indicate that the use of EEG-FM-based images as input to DarkNet19 architecture gives significant advantages in the detection of ADHD. [ABSTRACT FROM AUTHOR]

According to several studies, it is feasible to significantly raise the detection engine's effectiveness and accuracy by choosing the right features for a threat detection system. New advances like Distributed Computing and Enormous Information have expanded network traffic, and the danger identification framework must proactively gain and dissect the information delivered by the approaching traffic. Nonetheless, not all elements in an enormous dataset help to portray the traffic, therefore restricting and picking few reasonable highlights might accelerate and improve the danger discovery framework's exactness. Deep neural networks enhance the detection rates of intrusion detection models, making machine learning-based intrusion detection systems (IDS's) useful recently. Consumers, however, find it more and more challenging to comprehend the reasoning behind their selections as models become more complex accuracy. Using relevant features from the NSL-KDD dataset, we apply appropriate feature selection mechanisms to implement an intrusion detection system to implement a faster system with increased accuracy. We use Explainable Model (SHAP) to interpret the results of IDS. The interpretation of findings utilizing the Explainable Model (SHAP) for machine learning (ML) and deep learning (DL) models heavily depends on its efficiency. While DL models require more resources, ML models are computationally efficient. Both models, however, gain from SHAP interpretations, which offer perceptions into the significance of features and contributions to predictions. While DL models excel in accuracy, ML models offer efficiency. The decision is based on the particular needs and resources that are available, with SHAP offering greater knowledge of model behavior and feature impact. [ABSTRACT FROM AUTHOR]

Purpose: Diabetes is one of the life-threatening chronic diseases, which is already affecting 422m people globally based on (World Health Organization) WHO report as at 2018. This costs individuals, government and groups a whole lot; right from its diagnosis stage to the treatment stage. The reason for this cost, among others, is that it is a long-term treatment disease. This disease is likely to continue to affect more people because of its long asymptotic phase, which makes its early detection not feasible. Design/methodology/approach: In this study, the authors have presented machine learning models with feature selection, which can detect diabetes disease at its early stage. Also, the models presented are not costly and available to everyone, including those in the remote areas. Findings: The study result shows that feature selection helps in getting better model, as it prevents overfitting and removes redundant data. Hence, the study result when compared with previous research shows the better result has been achieved, after it was evaluated based on metrics such as F-measure, Precision-Recall curve and Receiver Operating Characteristic Area Under Curve. This discovery has the potential to impact on clinical practice, when health workers aim at diagnosing diabetes disease at its early stage. Originality/value: This study has not been published anywhere else. [ABSTRACT FROM AUTHOR]

Many studies now indicate that brain disorders are associated with functional connectivity between brain regions, but the impact of interactions among functional connections on diseases remains to be explored. Moreover, existing models for analyzing the temporal and spatial attributes of fMRI data exhibit limitations in their efficacy. Currently, there is an absence of a comprehensive model capable of simultaneously capturing disease-related interactions within functional connectivity, fully leveraging the temporal and spatial characteristics of fMRI, and achieving superior diagnostic accuracy. To address these gaps, we propose a model to classify brain disorders that employs a novel algorithm and multimodal graph convolutional networks to process the temporal and spatial features of fMRI data separately, along with the demographic information. In the proposed model, a deep feature selection algorithm is employed to discover disease-related biomarkers and evaluate their impact on brain disorders. Remarkably, this approach has led to a classification accuracy of 80.66%, which surpasses the performance of graph neural networks by approximately 10%. Notably, our investigations have uncovered novel biomarkers linked to autism, with our findings corroborated by relevant literature. These achievements underscore the model's potential in revealing critical insights pertinent to autism, thereby facilitating groundbreaking experimental research in the field. [ABSTRACT FROM AUTHOR]

Complex systems are prone to faults due to their intricate structures, potentially impacting system stability. Therefore, fault diagnosis has become crucial for maintaining stable operation. In the field of complex systems, the combinatorial explosion problem in belief rule base (BRB) has attracted significant attention. The interdependence among system components leads to numerous variables and the need for rules, heightening model complexity. Regarding the combinatorial explosion problem, an improved belief rule network structure called deep BRB (DBRB) is proposed. First, the extreme gradient boosting (XGBoost) feature selection method is employed to choose the relatively important feature subset. Next, driven by the importance of features, different levels of features are input into the model, forming a complete and progressive network structure. Finally, the model undergoes the reasoning and optimization process. The effectiveness of the model is confirmed with a bearing fault dataset. After a comprehensive evaluation of multiple indicators, this method demonstrates a consistent improvement in classification performance as the depth increased. Moreover, compared to the traditional BRB model, this method notably reduces the number of parameters, improving its efficiency of processing complex data. In short, this method effectively tackles combinatorial explosion while ensuring model performance. The selection and assignment of feature subsets enhance the logic and readability of the model. Through the network structure, various fault features are captured well. This fault diagnosis method, rooted in the DBRB, offers a novel perspective on diagnosing complex system faults. • A deep structure is proposed to solve the combinatorial explosion problem in fault diagnosis of complex systems. • Progressive fault diagnosis enhances the rationality and accuracy of diagnosis. • Compatibility is considered in the design of the method. [ABSTRACT FROM AUTHOR]

Pneumonia in Peru is a very serious problem. Its impact in recent years has been aggravated due to the Covid-19 pandemic, generating an increase in infections and deaths without distinguishing the age range, which placed this country on the mortality list due to the pandemic. That is why this research seeks the causes of this problem and evaluates what patterns were detected between the years 2019-2022 in patients with pneumonia in Peru from data set from the Comprehensive Health Insurance (SIS). The data presented values related to age, gender, medication and other significant values to understand the disease. The results of the research were achieved by using the PCA technique where the dimensionality of the data was reduced from 28 to 4 main features (Patient's year of health care, Age, BMI, Department). Finally, with this processed data set, the K-Means algorithm was used, where it was determined that patients in the 60 to 85 years range are the most affected by J189 pneumonia. In addition, an environmental pattern was found in J189 pneumonia. J128, resulting in a focus on patients on the Peruvian coast in places like Lima or La Libertad. [ABSTRACT FROM AUTHOR]

This study focuses on enhancing DoS attack detection in IoT systems through a Machine Learning approach that combines class balancing, feature selection, and optimized classifiers. Utilizing the Edge IIoT dataset, we applied SMOTE and Random Un-dersampling for class balance and employed DNN, Random Forest, and PCA for feature selection. We evaluated six technique combinations across four classifiers (DNN, SVM, XGBoost, and Random Forest), finding that certain combinations notably improve detection efficiency and accuracy. This research contributes to IoT security by offering an effective methodology for DoS attack detection, setting a foundation for further advances in IoT system protection against security threats. [ABSTRACT FROM AUTHOR]

This paper presents a novel network intrusion detection framework that combines convolutional recurrent neural networks (CRNN) and random forest (RF) models within a federated learning setting. The proposed approach aims to address the challenges of data privacy, computational efficiency, and model generalization in traditional network intrusion detection methods. By leveraging the spatial feature extraction capabilities of CRNN and the feature selection and noise reduction properties of RF, the framework enhances the accuracy and robustness of attack detection. The integration of federated learning enables collaborative model training without compromising data privacy. Extensive experiments on benchmark datasets demonstrate the superiority of the proposed method compared to state-of-the-art techniques, achieving high performance metrics such as accuracy, precision, recall, F1 score, and AUC. The proposed framework offers a promising solution for secure and efficient network intrusion detection in real-world scenarios, contributing to the advancement of cybersecurity practices. [ABSTRACT FROM AUTHOR]

Functional End-of-Line Testing is a powerful but expensive approach for industrial quality assurance. A major cost driver of End-of-Line Testing is the often overlong and overcomplex design of industrial test cycles. To tackle this challenge, the paper proposes a data-driven approach based on machine learning for the optimization of the End-of-Line test cycle of a hydraulic pump. As each test cycle comprises a multivariate and multi-phased time series, the optimization can be interpreted as a form of feature selection that aims to select the minimal subset of relevant and non-redundant testing sensors and cycle phases. We define the problem as a multi-objective optimization task balancing the goals of maximizing the test cycle sensitivity and specificity while minimizing the test cycle complexity. The general idea of the optimization approach is inspired by the iterative concept of wrapper feature selection. Our approach was validated in an experiment by applying it to real-world End-of-Line Testing data of an hydraulic axial piston pump. The results have shown that the approach could be successfully used to find an improved test cycle design for the pump and, especially, Backward Elimination performed well as solution search algorithm. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a progressive neurodegenerative illness triggered by decreased dopamine secretion. Fuzzy logic has gained substantial attention in PD diagnosis research. PD detection using fuzzy logic has presented more precise outcomes as compared with common machine learning approaches. In this research, a hybrid method combining supervised learning, unsupervised learning and feature selection techniques is developed. In a type-1 fuzzy system, the membership functions used for the fuzzification of the crisp inputs are mapped to single numbers. However, in a type-2 fuzzy system, these numbers are represented as intervals, adding an extra dimension to the definition of the membership function. The first step of the proposed method involves clustering the data using the Expectation–Maximization (EM) technique. The performance of EM clustering is performed using the Davies–Bouldin index. Subsequently, feature selection is performed using the backward stepwise regression. To predict the UPDRS, Type-2 Sugeno fuzzy inference system (T2SFIS) is implemented on the clusters generated from the previous steps. The Parkinson's telemonitoring dataset is used in this study for method evaluation. Using the EM algorithm, the PD dataset was clustered into 13 segments, and the most important features for accurate UPDRS prediction were chosen in each segment using backward stepwise regression. The hybrid method was evaluated using R-squared (R2) and RMSE. The evaluation results showed that the combination of EM, backward stepwise regression, and type-2 Sugeno FIS obtained the best accuracy in predicting Motor-UPDRS and Total-UPDRS. [ABSTRACT FROM AUTHOR]

Copyright of Informatica (03505596) is the property of Slovene Society Informatika and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This research proposes a novel approach for Tamil Handwritten Character Recognition (THCR) that combines feature selection and ensemble learning techniques. The Tamil script is complex and highly variable, requiring a robust and accurate recognition system. Feature selection is used to reduce dimensionality while preserving discriminative features, improving classification performance and reducing computational complexity. Several feature selection methods are compared, and individual classifiers (support vector machines, neural networks, and decision trees) are evaluated through extensive experiments. Ensemble learning techniques such as bagging, and boosting are employed to leverage the strengths of multiple classifiers and enhance recognition accuracy. The proposed approach is evaluated on the HP Labs Dataset, achieving an impressive 95.56% accuracy using an ensemble learning framework based on support vector machines. The dataset consists of 82,928 samples with 247 distinct classes, contributed by 500 participants from Tamil Nadu. It includes 40,000 characters with 500 user variations. The results surpass or rival existing methods, demonstrating the effectiveness of the approach. The research also offers insights for developing advanced recognition systems for other complex scripts. Future investigations could explore the integration of deep learning techniques and the extension of the proposed approach to other Indic scripts and languages, advancing the field of handwritten character recognition. [ABSTRACT FROM AUTHOR]

Structural damage detection (SDD) remains highly challenging, due to the difficulty in selecting the optimal damage features from a vast amount of information. In this study, a tree model-based method using decision tree and random forest was employed for feature selection of vibration response signals in SDD. Signal datasets were obtained by numerical experiments and vibration experiments, respectively. Dataset features extracted using this method were input into a convolutional neural network to determine the location of structural damage. Results indicated a 5% to 10% improvement in detection accuracy compared to using original datasets without feature selection, demonstrating the feasibility of this method. The proposed method, based on tree model and classification, addresses the issue of extracting effective information from numerous vibration response signals in structural health monitoring. [ABSTRACT FROM AUTHOR]