Expert Systems最新文献_第4页

Enhanced Electricity Forecasting for Smart Buildings Using a TCN-Bi-LSTM Deep Learning Model

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-30 DOI: 10.1111/exsy.70000

Sandeep Kumar Gautam, Vinayak Shrivastava, Sandeep S. Udmale

{"title":"Enhanced Electricity Forecasting for Smart Buildings Using a TCN-Bi-LSTM Deep Learning Model","authors":"Sandeep Kumar Gautam, Vinayak Shrivastava, Sandeep S. Udmale","doi":"10.1111/exsy.70000","DOIUrl":"https://doi.org/10.1111/exsy.70000","url":null,"abstract":"<div>\u0000 \u0000 <p>Integration of sensor technology and advanced software empowers consumers to manage energy usage proactively. This proactive approach yields positive impacts at both micro and macro levels, benefiting individuals and contributing to broader environmental conservation efforts. By leveraging predictive models, consumers can make informed decisions that serve their interests and promote a greener and more sustainable future for all. Thus, energy consumption (EC) prediction is crucial for effective resource management. In this study, we propose an innovative deep-learning approach to predict EC, focusing specifically on smart buildings. Our model utilises a hybrid deep learning architecture to effectively capture low and high information patterns present in multivariate time series data of various sensors deployed in smart buildings and numerous influencing factors. To address the nonlinear and dynamic nature of this data, our model combines a deep neural network (DNN) with a deep learning sequential model (DLS). Specifically, temporal convolutional networks (TCN) within the DNN family are employed to extract various trends from the data, while the DLS model, which consists of Bi-directional Long Short-term Memory Networks (Bi-LSTM), is employed to learn and capture these trends effectively. Consequently, we present a hybrid deep learning framework that leverages for learning multivariate time series data related to EC with shared feature representation. To validate our approach, we extensively evaluate our model using a dataset from an office building in Berkeley, California. Experimental results demonstrate that our model achieves satisfactory accuracy in EC prediction. For the 7-h horizon and on multivariate TS data, an <i>R</i><sup>2</sup> of 0.97 is realised for the proposed model. This is confirmed by the 1.65% improvement in transiting from univariate to multivariate data, which supports using multiple modalities.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143121052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Phishing Website Detection: An In-Depth Investigation of Feature Selection and Deep Learning

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-29 DOI: 10.1111/exsy.13824

Soudabe Mousavi, Mahdi Bahaghighat

{"title":"Phishing Website Detection: An In-Depth Investigation of Feature Selection and Deep Learning","authors":"Soudabe Mousavi, Mahdi Bahaghighat","doi":"10.1111/exsy.13824","DOIUrl":"https://doi.org/10.1111/exsy.13824","url":null,"abstract":"<div>\u0000 \u0000 <p>Cloud and fog computing technologies benefit from integrating AI-driven phishing detection as it enhances security, scalability, real-time reaction, and privacy. Nowadays, there is a noticeable rise in illegal activity taking place online. One of the illicit cybersecurity practices is phishing, in which hackers trick consumers by pretending to be authentic websites and spoofing them to obtain sensitive user information. Phishing attacks, regrettably, have increased dramatically in recent years, according to research. Machine learning (ML) and deep learning (DL) techniques have shown encouraging progress in thwarting these attacks. Consequently, we employed DL and ML techniques to identify phishing websites in this study. This article presents four scenarios in both ML and DL models. Two are proposed in ML, while the others are employed in DL. The outcomes of four scenarios were contrasted to determine which algorithm performed better at distinguishing between legal and illicit websites. Many popular ML techniques were used, including K-nearest neighbour, random forest (RF), decision trees, and SVMs. PCA and Importance Features are implemented in both ML scenarios to find the best features. RF successfully reached an accuracy of 97.82% using the Importance Feature technique. However, the PCA method failed to improve the performance of ML algorithms. As a result of ML-based scenarios, 98 features are selected for the final deep learning scenarios. In DL-based scenarios, algorithm architectures are essential to avoid overfitting and bias due to various hyperparameters. Thus, in the third scenario, our aim focuses on DL architecture design. Multilayer perceptron and convolutional neural networks (CNNs) are employed to detect phishing websites. Finally, our proposed 1D CNN model, using stratified k-fold cross-validation, outperformed the classical ML algorithm, achieving 98.94% accuracy and 0.99 AUC-ROC score in detecting phishing websites.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143120587","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ChartNet: Reducing Subjectivity in Stock Prediction Through Unified Technical Chart Representation

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-29 DOI: 10.1111/exsy.13841

Shangzhe Li, Yingke Liu, Fanglei Cheng, Junran Wu, Ke Xu

{"title":"ChartNet: Reducing Subjectivity in Stock Prediction Through Unified Technical Chart Representation","authors":"Shangzhe Li, Yingke Liu, Fanglei Cheng, Junran Wu, Ke Xu","doi":"10.1111/exsy.13841","DOIUrl":"https://doi.org/10.1111/exsy.13841","url":null,"abstract":"<div>\u0000 \u0000 <p>Technical analysis, which includes technical indicators and charts derived from specific rules, has proven effective and widely used for stock movement prediction. However, technical chart evaluation is often limited by subjectivity, arising from sparse chart types and substantial information loss due to rigid rules. While pattern recognition algorithms have been developed to address this issue, they still rely on manual chart labelling and primarily focus on closing prices, leaving much of the chart's broader information untapped. To overcome these limitations, we propose a novel framework called ChartNet, designed to extract general information from technical charts and reduce subjectivity in chart analysis. ChartNet employs a unified representation for charts across financial series with varying simplification levels and leverages a chart triplet loss function for unsupervised training, eliminating the need for labelled data. Compared with several state-of-the-art baselines, our framework has reached the best prediction accuracy on CSI-300, SZ-50 components and Dow Jones Index in 2022: 65.91%, 63.70% and 64.96% respectively. In backtesting using actual stock data, our framework achieves the highest average return of 1.12 and 1.15. Furthermore, we highlight the interpretability of ChartNet through two case studies, some important charts and failure cases, illustrating its capability to uncover meaningful insights from charts. This research contributes to advancing the objective evaluation of technical charts and promoting a more comprehensive understanding of chart-based stock prediction performance.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143120588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improving Safety and Efficiency of Industrial Vehicles by Bio-Inspired Algorithms

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-22 DOI: 10.1111/exsy.13836

Eduardo Bayona, J. Enrique Sierra-García, Matilde Santos Peñas

{"title":"Improving Safety and Efficiency of Industrial Vehicles by Bio-Inspired Algorithms","authors":"Eduardo Bayona, J. Enrique Sierra-García, Matilde Santos Peñas","doi":"10.1111/exsy.13836","DOIUrl":"https://doi.org/10.1111/exsy.13836","url":null,"abstract":"<div>\u0000 \u0000 <p>In the context of industrial automation, optimising automated guided vehicle (AGV) trajectories is crucial for enhancing operational efficiency and safety. They must travel in crowded work areas and cross narrow corridors with strict safety and time requirements. Bio-inspired optimization algorithms have emerged as a promising approach to deal with complex optimization scenarios. Thus, this paper explores the ability of three novel bio-inspired algorithms: the Bat Algorithm (BA), the Whale Optimization Algorithm (WOA) and the Gazelle Optimization Algorithm (GOA); to optimise the AGV path planning in complex environments. To do it, a new optimization strategy is described: the AGV trajectory is based on clothoid curves and a specialised piece-wise fitness function which prioritises safety and efficiency is designed. Simulation experiments were conducted across different occupancy maps to evaluate the performance of each algorithm. WOA demonstrates faster optimization providing suitable safety solutions 4 times faster than GOA. Meanwhile, GOA gives solutions with better safety metrics but demands more computational time. The study highlights the potential of bio-inspired approaches for AGV trajectory optimisation and suggests avenues for future research, including hybrid algorithm development.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143118131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unveiling the Differences in Early Performance Prediction Between Online Social Sciences and STEM Courses Using Educational Data Mining

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-20 DOI: 10.1111/exsy.13837

Raúl Marticorena-Sánchez, Antonio Canepa-Oneto, Carlos López-Nozal, José A. Barbero-Aparicio

{"title":"Unveiling the Differences in Early Performance Prediction Between Online Social Sciences and STEM Courses Using Educational Data Mining","authors":"Raúl Marticorena-Sánchez, Antonio Canepa-Oneto, Carlos López-Nozal, José A. Barbero-Aparicio","doi":"10.1111/exsy.13837","DOIUrl":"https://doi.org/10.1111/exsy.13837","url":null,"abstract":"<div>\u0000 \u0000 <p>Educational Data Mining and Learning Analytics in virtual environments can be used to diagnose student performance problems at an early stage. Information that is useful for guiding the decisions of teachers managing academic training, so that students can successfully complete their course. However, student interaction patterns may vary depending on the knowledge domain. Our aim is to design a framework applicable to online Social Sciences and STEM courses, recommending methods for building accurate early performance prediction models. A large-scale comparative study of the accuracy of multiple classifiers applied to classify the interaction logs of 32,593 students from 9 Social Sciences and 13 STEM courses is presented. Corroborating the results of other works, it was observed that high early performance prediction accuracy was obtained based on nothing other than student logs: accuracies of 0.75 in the 10th week, 0.80 in the 20th week, 0.85 in the 30th week and 0.90 in the 40th week. However, accuracy rates were observed to vary significantly, in relation to the classification algorithm and the knowledge domain (Social Sciences vs. STEM). These predictions are generally less accurate for Social Sciences compared to STEM courses, especially at the beginning of the course, with fewer differences observed in the final weeks. Additionally, this research identifies instances of low-accuracy outliers in the prediction of Social Sciences courses over time. These findings highlight the complex challenges and variations in early performance prediction across different domains in online education.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143117565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring Metaheuristic Optimization Algorithms in the Context of Textual Cyberharassment: A Systematic Review

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-13 DOI: 10.1111/exsy.13826

Fatima Shannaq, Mohammad Shehab, Areej Alshorman, Mahmoud Hammad, Bassam Hammo, Wala'a Al-Omari

{"title":"Exploring Metaheuristic Optimization Algorithms in the Context of Textual Cyberharassment: A Systematic Review","authors":"Fatima Shannaq, Mohammad Shehab, Areej Alshorman, Mahmoud Hammad, Bassam Hammo, Wala'a Al-Omari","doi":"10.1111/exsy.13826","DOIUrl":"https://doi.org/10.1111/exsy.13826","url":null,"abstract":"<div>\u0000 \u0000 <p>The digital landscape and rapid advancement of Information and Communication Technology have significantly increased social interactions, but it has also led to a rise in harmful behaviours such as offensive language, cyberbullying, and HS. Addressing online harassment is critical due to its severe consequences. This study offers a comprehensive evaluation of existing studies that employed metaheuristic optimization algorithms for detecting textual harassment content across social media platforms, highlighting their strengths and limitations. Using the PRISMA methodology, we reviewed and analysed 271 research papers, ultimately narrowing down the selection to 36 papers based on specific inclusion and exclusion criteria. By analysing key factors such as optimization techniques, feature engineering strategies, and dataset characteristics, we identify crucial trends and challenges in the field. Finally, we offer practical recommendations to improve the accuracy of predictive models, including adopting hybrid approaches, enhancing multilingual capabilities, and expanding models to operate effectively across various social media platforms.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 2","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143114741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimal Task Allocation and Sequencing for Flight Test Based on a Memetic Algorithm With Lexicographic Optimisation

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-12 DOI: 10.1111/exsy.13800

Bei Tian, Gang Xiao, Yu Shen, Xingwei Jiang

{"title":"Optimal Task Allocation and Sequencing for Flight Test Based on a Memetic Algorithm With Lexicographic Optimisation","authors":"Bei Tian, Gang Xiao, Yu Shen, Xingwei Jiang","doi":"10.1111/exsy.13800","DOIUrl":"https://doi.org/10.1111/exsy.13800","url":null,"abstract":"<div>\u0000 \u0000 <p>The flight test plays an important role in the development of an aircraft. Currently, with the increasing complexity and higher validation requirements for aircraft, there is a crucial need to generate high-quality flight test task schedules in an efficient way. This paper proposes a flight test task scheduling problem (FTTSP), which involves assigning suitable aircraft and executing the flight test tasks in a given order. Generally, the flight test duration (FTD) is the primary optimisation objective for the flight test task schedule, as it has a direct impact on aircraft development costs and the time to enter the market. In this study, the FTTSP not only considers FTD but also takes into account task transfer consumption (TTC). A mixed-integer linear programming mathematical model is first formulated to describe the FTTSP characteristics with the optimisation of the FTD and the TTC in a sequential manner. Then, a memetic algorithm with lexicographic optimisation (MALO) is proposed, which can efficiently obtain a high-quality solution and ensure that the most critical metric can be fully optimised. In MALO, a two-vector encoding and a task logic relationship repair mechanism based on the binary tree are established. An idle time insertion decoding method is designed to improve the aircraft utilisation rate. In addition to the selection, crossover and mutation operators, a local search operator is designed to enhance the solution quality. Finally, the full-scale test instances are generated for the FTTSP to evaluate the algorithm's performance. The numerical results demonstrate the effectiveness and competitiveness of the MALO in generating a high-quality schedule for flight test tasks.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 2","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143114581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing Security and Performance in Live VM Migration: A Machine Learning-Driven Framework With Selective Encryption for Enhanced Security and Performance in Cloud Computing Environments

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-09 DOI: 10.1111/exsy.13823

Raseena M. Haris, Mahmoud Barhamgi, Ahmed Badawy, Armstrong Nhlabatsi, Khaled M. Khan

{"title":"Enhancing Security and Performance in Live VM Migration: A Machine Learning-Driven Framework With Selective Encryption for Enhanced Security and Performance in Cloud Computing Environments","authors":"Raseena M. Haris, Mahmoud Barhamgi, Ahmed Badawy, Armstrong Nhlabatsi, Khaled M. Khan","doi":"10.1111/exsy.13823","DOIUrl":"https://doi.org/10.1111/exsy.13823","url":null,"abstract":"<p>Live virtual machine (LVM) migration is pivotal in cloud computing for its ability to seamlessly transfer virtual machines (VMs) between physical hosts, optimise resource utilisation, and enable uninterrupted service. However, concerns persist regarding safeguarding sensitive data during migration, particularly in critical sectors like healthcare, banking and military operations. Existing migration methods often compromise between performance and data security, prompting the need for a balanced solution. To address this, we propose a novel framework merging machine learning with selective encryption to fortify the pre-copy live migration process. Our approach intelligently predicts optimal migration times while selectively encrypting sensitive data, ensuring confidentiality and integrity without compromising performance. Rigorous experiments demonstrate its effectiveness, showcasing an average 51.82% reduction in downtime and an average 72.73% decrease in total migration time across diverse workloads. This integration of selective encryption not only bolsters security but also optimises migration metrics, presenting a robust solution for uninterrupted service delivery in critical cloud computing domains.</p>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 2","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/exsy.13823","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143113364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Novel Reciprocal Domain Adaptation Neural Network for Enhanced Diagnosis of Chronic Kidney Disease

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-09 DOI: 10.1111/exsy.13825

Saeed Iqbal, Adnan N. Qureshi, Musaed Alhussein, Khursheed Aurangzeb, Muhammad Zubair, Amir Hussain

{"title":"A Novel Reciprocal Domain Adaptation Neural Network for Enhanced Diagnosis of Chronic Kidney Disease","authors":"Saeed Iqbal, Adnan N. Qureshi, Musaed Alhussein, Khursheed Aurangzeb, Muhammad Zubair, Amir Hussain","doi":"10.1111/exsy.13825","DOIUrl":"https://doi.org/10.1111/exsy.13825","url":null,"abstract":"<div>\u0000 \u0000 <p>Chronic kidney disease (CKD) is a major global health concern caused mostly by high blood pressure and glucose levels. Detecting CKD early is critical for reducing its negative consequences since it can lead to increased mortality rates. With CKD's rising incidence expected to make it the fifth biggest cause of death by 2040, rapid advances in diagnostic approaches are required. This study presents the Reciprocal Domain Adaptation Network (RDAN) as a potential approach to the various issues of CKD diagnosis. RDAN is a neural network model that will help to traverse the complexity of CKD diagnosis by smoothly combining diverse data sets. RDAN consists of two critical units at its foundation: Mutual Model Adaptation (MMA) and Domain Model Learning. The MMA unit uses a powerful Global and Local Pyramid Pooling technique to extract rich features from a variety of data domains. Meanwhile, the DML unit uses semi-supervised domain-independent features combined with MMA features to improve representation learning. RDAN includes a reciprocal regularizer to promote cross-domain knowledge transfer, maximising feature representation for accurate CKD identification. An analysis of RDAN's performance on a variety of real-world datasets showed remarkable results in terms of accuracy (96.94%), precision (98.81%), recall (98.73%), F1-Score (98.88%), and area under the curve (AUC—99.35%). These results highlight the unmatched expertise of RDAN in managing data bias, domain changes, and privacy issues related to CKD diagnosis. Beyond statistical measures, RDAN's implications promise revolutionary breakthroughs in early CKD identification and subsequent therapeutic therapies. RDAN stands out as a groundbreaking method for diagnosing CKD. It delivers exceptional accuracy and can be seamlessly applied in various clinical environments.</p>\u0000 </div>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 2","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143113885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Machine Learning Algorithms to Address the Polarity and Stigma of Mental Health Disclosures on Instagram

IF 3 4区计算机科学

Expert Systems Pub Date : 2025-01-09 DOI: 10.1111/exsy.13832

Noemí Merayo, Alba Ayuso-Lanchares, Clara González-Sanguino

{"title":"Machine Learning Algorithms to Address the Polarity and Stigma of Mental Health Disclosures on Instagram","authors":"Noemí Merayo, Alba Ayuso-Lanchares, Clara González-Sanguino","doi":"10.1111/exsy.13832","DOIUrl":"https://doi.org/10.1111/exsy.13832","url":null,"abstract":"<p>This research explores the social response to disclosures and conversations about mental health on social media, which is a pioneering and innovative approach. Unlike previous studies, which focused predominantly on psychopathological aspects, this study explores how communities react to conversations about mental health on Instagram, one of the favourite social media platforms among young people, breaking new ground not only in the Spanish context, but also on a global scale, filling a gap in international research. The study created a novel corpus by collecting and labelling comments on Instagram posts related to celebrity mental health disclosures, categorising them by polarity (positive, negative, neutral) and stigma. Additionally, the research implements machine learning algorithms to detect stigma and polarity in mental health disclosures on Instagram. While traditional techniques like Support Vector Machine (SVM) and RF (Random Forest) displayed decent performance with lower computational loads, advanced deep learning and BERT (Bidirectional Encoder Representation from Transformers) algorithms achieved outstanding results. In fact, BERT models achieve around 96% accuracy in polarity and stigma detection, while deep learning models achieve 80% for polarity and 87% for stigma, very high accuracy metrics. This research contributes significantly to understanding the impact of mental health discussions on social media, offering insights that can reduce stigma and raise awareness. Artificial intelligence can be used for more responsible use of social media and effective management of mental health problems in digital environments.</p>","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":"42 2","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/exsy.13832","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143113886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0