Applied AI letters最新文献

{"title":"On Using the Shapley Value for Anomaly Localization: A Statistical Investigation","authors":"Rick S. Blum,&nbsp;Xubin Fang,&nbsp;Franziska Freytag","doi":"10.1002/ail2.70024","DOIUrl":"10.1002/ail2.70024","url":null,"abstract":"<p>Recent publications have suggested using the Shapley value for anomaly localization for sensor data systems. We use a reasonable statistical model for the classifiers required to compute the Shapley value to provide repeatable and rigorous analysis in the anomaly localization application. Then we provide a proof that using a single fixed term in the Shapley value calculation achieves a lower complexity anomaly localization test, with the same probability of error, as a test using the Shapley value in cases with independent observation. While it is impossible to test all possible cases numerically, we found this to be true in all the cases we tested with independent observations. For some dependent observation cases with two sensors, where only the second sensor data is anomalous, we show numerically that the Shapley value test can falsely decide an anomaly occurs at the first (nonanomalous) sensor with a probability which approaches one for increasing anomaly magnitude. On the other hand, using a single fixed term in the Shapley value calculation in these cases gives a reasonably small probability of an anomaly occurring at the first (nonanomalous) sensor for any anomaly magnitude. These results are the first of this type we have seen, could encourage new algorithm development, and should encourage future research to more fully understand these observations. A better understanding of the Shapley value, given its popularity, seems an important topic which could lead to improvements in algorithms and real implementations in the future.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70024","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147668494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Survey for Deep Reinforcement Learning Based Network Intrusion Detection","authors":"Wanrong Yang,&nbsp;Alberto Acuto,&nbsp;Yihang Zhou,&nbsp;Dominik Wojtczak","doi":"10.1002/ail2.70026","DOIUrl":"https://doi.org/10.1002/ail2.70026","url":null,"abstract":"<p>Cyber-attacks are gradually becoming more sophisticated and highly frequent nowadays, and the significance of network intrusion detection systems has become more pronounced. This paper investigates the prospects and challenges of employing deep reinforcement learning technologies in network intrusion detection. It begins with an introduction to the fundamental theories and technological frameworks of deep reinforcement learning including classic deep Q-network and actor-critic algorithms, followed by a review of essential research that has leveraged deep reinforcement learning for network intrusion detection in recent years. This research assesses these challenges and efforts in terms of model training efficiency, the detection capabilities for minority and unknown class attacks, improved network feature selection and unbalanced dataset issues. Performances of deep reinforcement learning models are comprehensively investigated. The findings reveal that although deep reinforcement learning shows promise in network intrusion detection, many of the latest deep reinforcement learning technologies are yet to be fully explored. Some deep reinforcement learning based models can achieve state-of-the-art results in some public datasets, in some cases, even better than traditional deep learning methods. The paper concludes with recommendations for the enhanced deployment and testing of deep reinforcement learning technologies in real-world network scenarios to further improve their application. Special emphasis is placed on the Internet of Things intrusion detection. We offer discussions on recently proposed deep architectures, revealing possible future policy functions used for deep reinforcement learning based network intrusion detection. In the end, we propose integrating deep reinforcement learning and broader generative methods and models to assist and further improve their performance. These advancements aim to address the current gaps and facilitate more robust and adaptive network intrusion detection systems.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70026","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147566485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Survey for Deep Reinforcement Learning Based Network Intrusion Detection","authors":"Wanrong Yang,&nbsp;Alberto Acuto,&nbsp;Yihang Zhou,&nbsp;Dominik Wojtczak","doi":"10.1002/ail2.70026","DOIUrl":"https://doi.org/10.1002/ail2.70026","url":null,"abstract":"<p>Cyber-attacks are gradually becoming more sophisticated and highly frequent nowadays, and the significance of network intrusion detection systems has become more pronounced. This paper investigates the prospects and challenges of employing deep reinforcement learning technologies in network intrusion detection. It begins with an introduction to the fundamental theories and technological frameworks of deep reinforcement learning including classic deep Q-network and actor-critic algorithms, followed by a review of essential research that has leveraged deep reinforcement learning for network intrusion detection in recent years. This research assesses these challenges and efforts in terms of model training efficiency, the detection capabilities for minority and unknown class attacks, improved network feature selection and unbalanced dataset issues. Performances of deep reinforcement learning models are comprehensively investigated. The findings reveal that although deep reinforcement learning shows promise in network intrusion detection, many of the latest deep reinforcement learning technologies are yet to be fully explored. Some deep reinforcement learning based models can achieve state-of-the-art results in some public datasets, in some cases, even better than traditional deep learning methods. The paper concludes with recommendations for the enhanced deployment and testing of deep reinforcement learning technologies in real-world network scenarios to further improve their application. Special emphasis is placed on the Internet of Things intrusion detection. We offer discussions on recently proposed deep architectures, revealing possible future policy functions used for deep reinforcement learning based network intrusion detection. In the end, we propose integrating deep reinforcement learning and broader generative methods and models to assist and further improve their performance. These advancements aim to address the current gaps and facilitate more robust and adaptive network intrusion detection systems.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70026","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147566521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"User-Based Evaluation of Explainability Techniques for Misogyny Detection in Code-Mixed Hindi–English","authors":"Sargam Yadav,&nbsp;Abhishek Kaushik,&nbsp;Kevin McDaid","doi":"10.1002/ail2.70028","DOIUrl":"https://doi.org/10.1002/ail2.70028","url":null,"abstract":"<p>Although artificial intelligence (AI) models have demonstrated great success and efficiency in automatically countering online hate speech and misogyny, they suffer from a lack of explainability and transparency. Explainable artificial intelligence (XAI) is an apparent solution to make opaque black-box models more explainable, but remains under-explored for low resource and code-mixed languages. In this study, user-based evaluation of 3 explainability techniques for a misogyny classifier has been performed through simulatability, what-if explainability, and a qualitative feedback that covers several key dimensions. The participants were also required to mark the rationales to justify their predictions, and required to answer an open-ended question regarding their opinion on the explanations. Analysis of the results from 10 participants highlights that Integrated Gradients (IG) was more consistent and plausible than other techniques, and their preference for token attribution heatmaps and bar plots over natural language explanations (NLE).</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70028","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"User-Based Evaluation of Explainability Techniques for Misogyny Detection in Code-Mixed Hindi–English","authors":"Sargam Yadav,&nbsp;Abhishek Kaushik,&nbsp;Kevin McDaid","doi":"10.1002/ail2.70028","DOIUrl":"https://doi.org/10.1002/ail2.70028","url":null,"abstract":"<p>Although artificial intelligence (AI) models have demonstrated great success and efficiency in automatically countering online hate speech and misogyny, they suffer from a lack of explainability and transparency. Explainable artificial intelligence (XAI) is an apparent solution to make opaque black-box models more explainable, but remains under-explored for low resource and code-mixed languages. In this study, user-based evaluation of 3 explainability techniques for a misogyny classifier has been performed through simulatability, what-if explainability, and a qualitative feedback that covers several key dimensions. The participants were also required to mark the rationales to justify their predictions, and required to answer an open-ended question regarding their opinion on the explanations. Analysis of the results from 10 participants highlights that Integrated Gradients (IG) was more consistent and plausible than other techniques, and their preference for token attribution heatmaps and bar plots over natural language explanations (NLE).</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70028","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147565031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Student Perspectives on GenAI: Usage, Attitudes, and Academic Integrity in Psychology Education","authors":"Anthony Bloxham,&nbsp;Thomas Newham,&nbsp;Chloe Ryding,&nbsp;Lee Hadlington","doi":"10.1002/ail2.70025","DOIUrl":"https://doi.org/10.1002/ail2.70025","url":null,"abstract":"<p>Generative artificial intelligence (GenAI) raises pressing pedagogical and ethical questions in higher education. We surveyed 87 UK psychology students about GenAI familiarity, study uses, attitudes, and the justification of questionable uses (neutralisation). 54% reported using GenAI to assist their studies, primarily via ChatGPT. Compared with non-users, study users showed more positive AI attitudes and higher neutralisation scores. Across the full sample, AI attitudes modestly predicted neutralisation. The most common study uses were explaining concepts and generating ideas, and most users intended to use GenAI again. Non-users were more likely to endorse restrictive views on GenAI in assessed work. Findings point to a tension between perceived learning value and risks of dependency and academic integrity. Students also reported a need for clearer institutional guidance. We recommend a balanced approach that supports responsible use, feedback literacy, and critical engagement with AI outputs, alongside continued student-centred research to inform policy and assessment design.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147565030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Student Perspectives on GenAI: Usage, Attitudes, and Academic Integrity in Psychology Education","authors":"Anthony Bloxham,&nbsp;Thomas Newham,&nbsp;Chloe Ryding,&nbsp;Lee Hadlington","doi":"10.1002/ail2.70025","DOIUrl":"https://doi.org/10.1002/ail2.70025","url":null,"abstract":"<p>Generative artificial intelligence (GenAI) raises pressing pedagogical and ethical questions in higher education. We surveyed 87 UK psychology students about GenAI familiarity, study uses, attitudes, and the justification of questionable uses (neutralisation). 54% reported using GenAI to assist their studies, primarily via ChatGPT. Compared with non-users, study users showed more positive AI attitudes and higher neutralisation scores. Across the full sample, AI attitudes modestly predicted neutralisation. The most common study uses were explaining concepts and generating ideas, and most users intended to use GenAI again. Non-users were more likely to endorse restrictive views on GenAI in assessed work. Findings point to a tension between perceived learning value and risks of dependency and academic integrity. Students also reported a need for clearer institutional guidance. We recommend a balanced approach that supports responsible use, feedback literacy, and critical engagement with AI outputs, alongside continued student-centred research to inform policy and assessment design.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Topological Graph Neural Networks: A Novel Approach for Geometric Deep Learning","authors":"Amarjeet,&nbsp;Anurag Sinha,&nbsp;Khalid Hussain,&nbsp;Wasim Ahmad Bhat,&nbsp;Sudhanshu Maurya,&nbsp;Mohammad Nadeem Ahmed,&nbsp;Mohammad Rashid Hussain,&nbsp;S. Khalid","doi":"10.1002/ail2.70021","DOIUrl":"https://doi.org/10.1002/ail2.70021","url":null,"abstract":"<p>This paper introduces a novel approach to graph neural networks (GNNs) that incorporates topological data analysis (TDA) to enhance the representational power of traditional GNN architectures. We propose <i>Topological Graph Neural Networks</i> (TopGNNs), which leverage persistent homology and simplicial complexes to capture multi-scale structural information that conventional GNNs often overlook. Our experimental results on benchmark datasets demonstrate that TopGNNs achieve competitive performance compared to state-of-the-art methods, particularly for tasks requiring sensitivity to global graph structure. We provide a comprehensive empirical analysis of TopGNNs across various domains including molecular property prediction, social network analysis, and citation networks. The proposed framework bridges the gap between algebraic topology and deep learning on graphs, offering a promising direction for future research in geometric deep learning.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70021","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147565029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Topological Graph Neural Networks: A Novel Approach for Geometric Deep Learning","authors":"Amarjeet,&nbsp;Anurag Sinha,&nbsp;Khalid Hussain,&nbsp;Wasim Ahmad Bhat,&nbsp;Sudhanshu Maurya,&nbsp;Mohammad Nadeem Ahmed,&nbsp;Mohammad Rashid Hussain,&nbsp;S. Khalid","doi":"10.1002/ail2.70021","DOIUrl":"https://doi.org/10.1002/ail2.70021","url":null,"abstract":"<p>This paper introduces a novel approach to graph neural networks (GNNs) that incorporates topological data analysis (TDA) to enhance the representational power of traditional GNN architectures. We propose <i>Topological Graph Neural Networks</i> (TopGNNs), which leverage persistent homology and simplicial complexes to capture multi-scale structural information that conventional GNNs often overlook. Our experimental results on benchmark datasets demonstrate that TopGNNs achieve competitive performance compared to state-of-the-art methods, particularly for tasks requiring sensitivity to global graph structure. We provide a comprehensive empirical analysis of TopGNNs across various domains including molecular property prediction, social network analysis, and citation networks. The proposed framework bridges the gap between algebraic topology and deep learning on graphs, offering a promising direction for future research in geometric deep learning.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70021","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unveiling Marketing Competencies for MBA Graduates Through Text Mining and Machine Learning Applications","authors":"Latha Arumugham,&nbsp;Mohanamani Palanisamy,&nbsp;Maria Tresita Paul Vincent,&nbsp;Md Billal Hossain","doi":"10.1002/ail2.70020","DOIUrl":"https://doi.org/10.1002/ail2.70020","url":null,"abstract":"<p>In today's dynamic business environment, marketing professionals are expected to demonstrate a balanced mix of conceptual knowledge, technical skills, and behavioral competencies while addressing their customers. This study analyses 2000 job postings from job web portal, collected between April and June 2023, to identify the competencies most demanded from master's in business administration (MBA) graduates specializing in marketing. Using a text mining approach, including systematic preprocessing, bigram extraction, and term frequency-inverse document frequency (TF-IDF) based term weighting, the extracted competencies were categorized into knowledge, skill, and behavioral based competencies. Findings of the study reveal strong emphasis by employers on marketing research, marketing management, and sales management as core knowledge-based competencies required. While selling ability, client relationship management, and digital marketing have evolved as key skill-based competencies. Time management, teamwork, and leadership evolved as critical behavioral attributes needed for an MBA graduate specializing in the marketing domain. These insights highlight the need for management curricula to integrate conceptual learning with practical skill development and structured behavioral guidance. The results also provide accreditation bodies with evidence-based competency trends for strengthening outcome-based frameworks and guiding industry partners in aligning recruitment, training, and academic collaboration efforts. By offering data driven competencies grounded in real job market demand, the study supports the development of more industry ready MBA marketing graduates.</p>","PeriodicalId":72253,"journal":{"name":"Applied AI letters","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ail2.70020","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147563825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}