Human Activity Recognition (HAR) is a highly promising research area meant to automatically identify and interpret human behaviour using data received from sensors in various contexts. The potential uses of HAR are many, among them health care, sports coaching or monitoring the elderly or disabled. Nonetheless, there are numerous hurdles to be circumvented for HAR's precision and usefulness to be improved. One of the challenges is that there is no uniformity in data collection and annotation making it difficult to compare findings among different studies. Furthermore, more comprehensive datasets are necessary so as to include a wider range of human activities in different contexts while complex activities, which consist of multiple sub‐activities, are still a challenge for recognition systems. Researchers have proposed new frontiers such as multi‐modal sensor data fusion and deep learning approaches for enhancing HAR accuracy while addressing these issues. Also, we are seeing more non‐traditional applications such as robotics and virtual reality/augmented world going forward with their use cases of HAR. This article offers an extensive review on the recent advances in HAR and highlights the major challenges facing this field as well as future opportunities for further researches.
人类活动识别（HAR）是一个极具发展前景的研究领域，其目的是在各种情况下利用传感器接收到的数据自动识别和解释人类行为。人类活动识别（HAR）的潜在用途很多，其中包括医疗保健、体育指导或监测老年人或残疾人。然而，要提高 HAR 的精确度和实用性，还有许多障碍需要克服。挑战之一是数据收集和注释不统一，因此很难比较不同研究的结果。此外，还需要更全面的数据集，以包括不同环境中更广泛的人类活动，而由多个子活动组成的复杂活动对识别系统来说仍是一个挑战。研究人员提出了一些新的前沿技术，如多模态传感器数据融合和深度学习方法，以便在解决这些问题的同时提高 HAR 的准确性。此外，我们还看到更多的非传统应用，如机器人和虚拟现实/增强世界，都在使用 HAR。本文对 HAR 的最新进展进行了广泛综述，并重点介绍了该领域面临的主要挑战以及未来进一步研究的机遇。
{"title":"Human activity recognition: A comprehensive review","authors":"Harmandeep Kaur, Veenu Rani, Munish Kumar","doi":"10.1111/exsy.13680","DOIUrl":"https://doi.org/10.1111/exsy.13680","url":null,"abstract":"Human Activity Recognition (HAR) is a highly promising research area meant to automatically identify and interpret human behaviour using data received from sensors in various contexts. The potential uses of HAR are many, among them health care, sports coaching or monitoring the elderly or disabled. Nonetheless, there are numerous hurdles to be circumvented for HAR's precision and usefulness to be improved. One of the challenges is that there is no uniformity in data collection and annotation making it difficult to compare findings among different studies. Furthermore, more comprehensive datasets are necessary so as to include a wider range of human activities in different contexts while complex activities, which consist of multiple sub‐activities, are still a challenge for recognition systems. Researchers have proposed new frontiers such as multi‐modal sensor data fusion and deep learning approaches for enhancing HAR accuracy while addressing these issues. Also, we are seeing more non‐traditional applications such as robotics and virtual reality/augmented world going forward with their use cases of HAR. This article offers an extensive review on the recent advances in HAR and highlights the major challenges facing this field as well as future opportunities for further researches.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141783662","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}
Afonso Baldo, Paulo J. S. Ferreira, João Mendes‐Moreira
With technological advancements, much data is being captured by sensors, smartphones, wearable devices, and so forth. These vast datasets are stored in data centres and utilized to forge data‐driven models for the condition monitoring of infrastructures and systems through future data mining tasks. However, these datasets often surpass the processing capabilities of traditional information systems and methodologies due to their significant size. Additionally, not all samples within these datasets contribute valuable information during the model training phase, leading to inefficiencies. The processing and training of Machine Learning algorithms become time‐consuming, and storing all the data demands excessive space, contributing to the Big Data challenge. In this paper, we propose two novel techniques to reduce large time‐series datasets into more compact versions without undermining the predictive performance of the resulting models. These methods also aim to decrease the time required for training the models and the storage space needed for the condensed datasets. We evaluated our techniques on five public datasets, employing three Machine Learning algorithms: Holt‐Winters, SARIMA, and LSTM. The outcomes indicate that for most of the datasets examined, our techniques maintain, and in several instances enhance, the forecasting accuracy of the models. Moreover, we significantly reduced the time required to train the Machine Learning algorithms employed.
{"title":"Sampling approaches to reduce very frequent seasonal time series","authors":"Afonso Baldo, Paulo J. S. Ferreira, João Mendes‐Moreira","doi":"10.1111/exsy.13690","DOIUrl":"https://doi.org/10.1111/exsy.13690","url":null,"abstract":"With technological advancements, much data is being captured by sensors, smartphones, wearable devices, and so forth. These vast datasets are stored in data centres and utilized to forge data‐driven models for the condition monitoring of infrastructures and systems through future data mining tasks. However, these datasets often surpass the processing capabilities of traditional information systems and methodologies due to their significant size. Additionally, not all samples within these datasets contribute valuable information during the model training phase, leading to inefficiencies. The processing and training of Machine Learning algorithms become time‐consuming, and storing all the data demands excessive space, contributing to the Big Data challenge. In this paper, we propose two novel techniques to reduce large time‐series datasets into more compact versions without undermining the predictive performance of the resulting models. These methods also aim to decrease the time required for training the models and the storage space needed for the condensed datasets. We evaluated our techniques on five public datasets, employing three Machine Learning algorithms: Holt‐Winters, SARIMA, and LSTM. The outcomes indicate that for most of the datasets examined, our techniques maintain, and in several instances enhance, the forecasting accuracy of the models. Moreover, we significantly reduced the time required to train the Machine Learning algorithms employed.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141783665","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}
Kyung‐yeul Kim, Young‐bo Yang, Mi‐ra Kim, Jihie Kim, Ji Su Park
When stress causes negative behaviours to emerge in our daily lives, it is important to intervene quickly and appropriately to control the negative problem behaviours. Questionnaires, a common method of information gathering, have the disadvantage that it is difficult to get the exact information needed due to defensive or insincere responses from subjects. As an alternative to these drawbacks, projective testing using pictures can provide the necessary information more accurately than questionnaires because the subject responds subconsciously and the direct experience expressed through pictures can be more accurate than questionnaires. Analysing hand‐drawn image data with the Wartegg Zeichen Test (WZT) is not easy. In this study, we used deep learning to analyse image data represented as pictures through WZT to predict early depression. We analyse the data of 54 people who were judged as early depression and 54 people without depression, and increase the number of people without depression to 100 and 500, and aim to study in unbalanced data. We use CNN and CNN‐SVM to analyse the drawing images of WZT's initial depression with deep learning and predict the outcome of depression. The results show that the initial depression is predicted with 92%–98% accuracy on the image data directly drawn by WZT. This is the first study to automatically analyse and predict early depression in WZT based on hand‐drawn image data using deep learning models. The extraction of features from WZT images by deep learning analysis is expected to create more research opportunities through the convergence of psychotherapy and Information and Communication Technology (ICT) technology, and is expected to have high growth potential.
{"title":"Predicting early depression in WZT drawing image based on deep learning","authors":"Kyung‐yeul Kim, Young‐bo Yang, Mi‐ra Kim, Jihie Kim, Ji Su Park","doi":"10.1111/exsy.13675","DOIUrl":"https://doi.org/10.1111/exsy.13675","url":null,"abstract":"When stress causes negative behaviours to emerge in our daily lives, it is important to intervene quickly and appropriately to control the negative problem behaviours. Questionnaires, a common method of information gathering, have the disadvantage that it is difficult to get the exact information needed due to defensive or insincere responses from subjects. As an alternative to these drawbacks, projective testing using pictures can provide the necessary information more accurately than questionnaires because the subject responds subconsciously and the direct experience expressed through pictures can be more accurate than questionnaires. Analysing hand‐drawn image data with the Wartegg Zeichen Test (WZT) is not easy. In this study, we used deep learning to analyse image data represented as pictures through WZT to predict early depression. We analyse the data of 54 people who were judged as early depression and 54 people without depression, and increase the number of people without depression to 100 and 500, and aim to study in unbalanced data. We use CNN and CNN‐SVM to analyse the drawing images of WZT's initial depression with deep learning and predict the outcome of depression. The results show that the initial depression is predicted with 92%–98% accuracy on the image data directly drawn by WZT. This is the first study to automatically analyse and predict early depression in WZT based on hand‐drawn image data using deep learning models. The extraction of features from WZT images by deep learning analysis is expected to create more research opportunities through the convergence of psychotherapy and Information and Communication Technology (ICT) technology, and is expected to have high growth potential.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141783664","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}
Shumaila Khan, Iqbal Qasim, Wahab Khan, Khursheed Aurangzeb, Javed Ali Khan, Muhammad Shahid Anwar
Sarcasm detection is challenging in natural language processing (NLP) due to its implicit nature, particularly in low‐resource languages. Despite limited linguistic resources, researchers have focused on detecting sarcasm on social media platforms, leading to the development of specialized algorithms and models tailored for Urdu text. Researchers have significantly improved sarcasm detection accuracy by analysing patterns and linguistic cues unique to the language, thereby advancing NLP capabilities in low‐resource languages and facilitating better communication within diverse online communities. This work introduces UrduSarcasmNet, a novel architecture using cascaded group multi‐head attention, which is an innovative deep‐learning approach that employs cascaded group multi‐head attention techniques to enhance effectiveness. By employing a series of attention heads in a cascading manner, our model captures both local and global contexts, facilitating a more comprehensive understanding of the text. Adding a group attention mechanism enables simultaneous consideration of various sub‐topics within the content, thereby enriching the model's effectiveness. The proposed UrduSarcasmNet approach is validated with the Urdu‐sarcastic‐tweets‐dataset (UST) dataset, which has been curated for this purpose. Our experimental results on the UST dataset show that the proposed UrduSarcasmNet framework outperforms the simple‐attention mechanism and other state‐of‐the‐art models. This research significantly enhances natural language processing (NLP) and provides valuable insights for improving sarcasm recognition tools in low‐resource languages like Urdu.
{"title":"A novel transformer attention‐based approach for sarcasm detection","authors":"Shumaila Khan, Iqbal Qasim, Wahab Khan, Khursheed Aurangzeb, Javed Ali Khan, Muhammad Shahid Anwar","doi":"10.1111/exsy.13686","DOIUrl":"https://doi.org/10.1111/exsy.13686","url":null,"abstract":"Sarcasm detection is challenging in natural language processing (NLP) due to its implicit nature, particularly in low‐resource languages. Despite limited linguistic resources, researchers have focused on detecting sarcasm on social media platforms, leading to the development of specialized algorithms and models tailored for Urdu text. Researchers have significantly improved sarcasm detection accuracy by analysing patterns and linguistic cues unique to the language, thereby advancing NLP capabilities in low‐resource languages and facilitating better communication within diverse online communities. This work introduces UrduSarcasmNet, a novel architecture using cascaded group multi‐head attention, which is an innovative deep‐learning approach that employs cascaded group multi‐head attention techniques to enhance effectiveness. By employing a series of attention heads in a cascading manner, our model captures both local and global contexts, facilitating a more comprehensive understanding of the text. Adding a group attention mechanism enables simultaneous consideration of various sub‐topics within the content, thereby enriching the model's effectiveness. The proposed UrduSarcasmNet approach is validated with the Urdu‐sarcastic‐tweets‐dataset (UST) dataset, which has been curated for this purpose. Our experimental results on the UST dataset show that the proposed UrduSarcasmNet framework outperforms the simple‐attention mechanism and other state‐of‐the‐art models. This research significantly enhances natural language processing (NLP) and provides valuable insights for improving sarcasm recognition tools in low‐resource languages like Urdu.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141783666","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}
Ezaz Mohammed Al‐dahasi, Rama Khaled Alsheikh, Fakhri Alam Khan, Gwanggil Jeon
The rapid advancement of the Internet and digital payments has transformed the landscape of financial transactions, leading to both technological progress and an alarming rise in cybercrime. This study addresses the critical issue of financial fraud detection in the era of digital payments, focusing on enhancing operational risk frameworks to mitigate the increasing threats. The objective is to improve the predictive performance of fraud detection systems using machine learning techniques. The methodology involves a comprehensive data preprocessing and model creation process, including one‐hot encoding, feature selection, sampling, standardization, and tokenization. Six machine learning models are employed for fraud detection, and their hyperparameters are optimized. Evaluation metrics such as accuracy, precision, recall, and F1‐score are used to assess model performance. Results reveal that XGBoost and Random Forest outperform other models, achieving a balance between false positives and false negatives. The study meets the requirements for fraud detection systems, ensuring accuracy, scalability, adaptability, and explainability. This paper provides valuable insights into the efficacy of machine learning models for financial fraud detection and emphasizes the importance of striking a balance between false positives and false negatives.
互联网和数字支付的快速发展改变了金融交易的格局，既带来了技术进步，也导致网络犯罪的惊人增长。本研究探讨了数字支付时代金融欺诈检测的关键问题，重点是加强操作风险框架，以减轻日益增长的威胁。目的是利用机器学习技术提高欺诈检测系统的预测性能。该方法涉及全面的数据预处理和模型创建过程，包括单次编码、特征选择、采样、标准化和标记化。欺诈检测采用了六个机器学习模型，并对其超参数进行了优化。准确率、精确度、召回率和 F1 分数等评价指标用于评估模型性能。结果显示，XGBoost 和随机森林的表现优于其他模型，在误报和误报之间取得了平衡。这项研究符合欺诈检测系统的要求，确保了准确性、可扩展性、适应性和可解释性。本文就机器学习模型在金融欺诈检测中的功效提供了宝贵的见解，并强调了在误报和误报之间取得平衡的重要性。
{"title":"Optimizing fraud detection in financial transactions with machine learning and imbalance mitigation","authors":"Ezaz Mohammed Al‐dahasi, Rama Khaled Alsheikh, Fakhri Alam Khan, Gwanggil Jeon","doi":"10.1111/exsy.13682","DOIUrl":"https://doi.org/10.1111/exsy.13682","url":null,"abstract":"The rapid advancement of the Internet and digital payments has transformed the landscape of financial transactions, leading to both technological progress and an alarming rise in cybercrime. This study addresses the critical issue of financial fraud detection in the era of digital payments, focusing on enhancing operational risk frameworks to mitigate the increasing threats. The objective is to improve the predictive performance of fraud detection systems using machine learning techniques. The methodology involves a comprehensive data preprocessing and model creation process, including one‐hot encoding, feature selection, sampling, standardization, and tokenization. Six machine learning models are employed for fraud detection, and their hyperparameters are optimized. Evaluation metrics such as accuracy, precision, recall, and <jats:italic>F</jats:italic>1‐score are used to assess model performance. Results reveal that XGBoost and Random Forest outperform other models, achieving a balance between false positives and false negatives. The study meets the requirements for fraud detection systems, ensuring accuracy, scalability, adaptability, and explainability. This paper provides valuable insights into the efficacy of machine learning models for financial fraud detection and emphasizes the importance of striking a balance between false positives and false negatives.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141783668","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}
As digital transformation accelerates, data generated in a convergence information and communication technology (ICT) environment must be secured. This data includes confidential information such as personal and financial information, so attackers spread malware in convergence ICT environments to steal this information. To protect convergence ICT environments from diverse cyber threats, deep learning models have been utilized for malware detection. However, accurately detecting rapidly generated variants and obfuscated malware is challenging. This study proposes a three‐tier malware detection (TMaD) scheme that utilizes a cloud‐fog‐edge collaborative architecture to analyse multi‐view features of executable files and detect malware. TMaD performs signature‐based malware detection at the edge device tier, then sends executables detected as unknown or benign to the fog tier. The fog tier conducts static analysis on non‐obfuscated executables and those transferred from the previous tier to detect variant malware. Subsequently, TMaD sends executables detected as benign in the fog tier to the cloud tier, where dynamic analysis is performed on obfuscated executables and those detected as benign to identify obfuscated malware. An evaluation of TMaD's detection performance resulted in an accuracy of 94.78%, a recall of 0.9794, a precision of 0.9535, and an f1‐score of 0.9663. This performance demonstrates that TMaD, by analysing executables across several tiers and minimizing false negatives, exhibits superior detection performance compared to existing malware detection models.
{"title":"TMaD: Three‐tier malware detection using multi‐view feature for secure convergence ICT environments","authors":"Jueun Jeon, Byeonghui Jeong, Seungyeon Baek, Young‐Sik Jeong","doi":"10.1111/exsy.13684","DOIUrl":"https://doi.org/10.1111/exsy.13684","url":null,"abstract":"As digital transformation accelerates, data generated in a convergence information and communication technology (ICT) environment must be secured. This data includes confidential information such as personal and financial information, so attackers spread malware in convergence ICT environments to steal this information. To protect convergence ICT environments from diverse cyber threats, deep learning models have been utilized for malware detection. However, accurately detecting rapidly generated variants and obfuscated malware is challenging. This study proposes a three‐tier malware detection (TMaD) scheme that utilizes a cloud‐fog‐edge collaborative architecture to analyse multi‐view features of executable files and detect malware. TMaD performs signature‐based malware detection at the edge device tier, then sends executables detected as unknown or benign to the fog tier. The fog tier conducts static analysis on non‐obfuscated executables and those transferred from the previous tier to detect variant malware. Subsequently, TMaD sends executables detected as benign in the fog tier to the cloud tier, where dynamic analysis is performed on obfuscated executables and those detected as benign to identify obfuscated malware. An evaluation of TMaD's detection performance resulted in an accuracy of 94.78%, a recall of 0.9794, a precision of 0.9535, and an f1‐score of 0.9663. This performance demonstrates that TMaD, by analysing executables across several tiers and minimizing false negatives, exhibits superior detection performance compared to existing malware detection models.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141742191","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}
Moolchand Sharma, Aryan Bhatia, Akhil, A. Dutta, Shtwai Alsubai
In the realm of Drug‐Target Interaction (DTI) prediction, this research investigates and contrasts the efficacy of diverse evolutionary algorithms in fine‐tuning a sophisticated hybrid deep learning model. Recognizing the critical role of DTI in drug discovery and repositioning, we tackle the challenges of binary classification by reframing the problem as a regression task. Our focus lies on the Convolution Self‐Attention Network with Attention‐based bidirectional Long Short‐Term Memory Network (CSAN‐BiLSTM‐Att), a hybrid model combining convolutional neural network (CNN) blocks, self‐attention mechanisms, and bidirectional LSTM layers. To optimize this complex model, we employ Differential Evolution (DE), Particle Swarm Optimization (PSO), Memetic Particle Swarm Optimization Algorithm (MPSOA), Fire Hawk Optimization (FHO), and Artificial Hummingbird Algorithm (AHA). Through thorough comparative analysis, we evaluate the performance of these evolutionary algorithms in enhancing the CSAN‐BiLSTM‐Att model's effectiveness. By examining the strengths and weaknesses of each algorithm, our study aims to provide valuable insights into DTI prediction, identifying the most effective evolutionary algorithm for hyperparameter tuning in advanced deep learning models. Notably, Fire‐hawk optimization (FHO) emerges as particularly promising, achieving the highest Concordance Index (C‐index) as 0.974 for KIBA datasets and 0.894 for DAVIS datasets and demonstrating exceptional accuracy in ranking continuous predictions across both the datasets.
在药物-目标相互作用（DTI）预测领域，本研究调查并对比了各种进化算法在微调复杂的混合深度学习模型方面的功效。认识到 DTI 在药物发现和重新定位中的关键作用，我们通过将问题重构为回归任务来应对二元分类的挑战。我们的重点是卷积自注意力网络与基于注意力的双向长短期记忆网络（CSAN-BiLSTM-Att），这是一个结合了卷积神经网络（CNN）块、自注意力机制和双向 LSTM 层的混合模型。为了优化这一复杂模型，我们采用了差分进化（DE）、粒子群优化（PSO）、记忆粒子群优化算法（MPSOA）、火鹰优化（FHO）和人工蜂鸟算法（AHA）。通过全面的比较分析，我们评估了这些进化算法在提高 CSAN-BiLSTM-Att 模型有效性方面的性能。通过研究每种算法的优缺点，我们的研究旨在为 DTI 预测提供有价值的见解，为高级深度学习模型的超参数调整确定最有效的进化算法。值得注意的是，火鹰优化（FHO）特别有前途，它在 KIBA 数据集和 DAVIS 数据集上分别获得了 0.974 和 0.894 的最高一致性指数（C-index），并在这两个数据集的连续预测排名中表现出了非凡的准确性。
{"title":"Optimizing hybrid deep learning models for drug‐target interaction prediction: A comparative analysis of evolutionary algorithms","authors":"Moolchand Sharma, Aryan Bhatia, Akhil, A. Dutta, Shtwai Alsubai","doi":"10.1111/exsy.13683","DOIUrl":"https://doi.org/10.1111/exsy.13683","url":null,"abstract":"In the realm of Drug‐Target Interaction (DTI) prediction, this research investigates and contrasts the efficacy of diverse evolutionary algorithms in fine‐tuning a sophisticated hybrid deep learning model. Recognizing the critical role of DTI in drug discovery and repositioning, we tackle the challenges of binary classification by reframing the problem as a regression task. Our focus lies on the Convolution Self‐Attention Network with Attention‐based bidirectional Long Short‐Term Memory Network (CSAN‐BiLSTM‐Att), a hybrid model combining convolutional neural network (CNN) blocks, self‐attention mechanisms, and bidirectional LSTM layers. To optimize this complex model, we employ Differential Evolution (DE), Particle Swarm Optimization (PSO), Memetic Particle Swarm Optimization Algorithm (MPSOA), Fire Hawk Optimization (FHO), and Artificial Hummingbird Algorithm (AHA). Through thorough comparative analysis, we evaluate the performance of these evolutionary algorithms in enhancing the CSAN‐BiLSTM‐Att model's effectiveness. By examining the strengths and weaknesses of each algorithm, our study aims to provide valuable insights into DTI prediction, identifying the most effective evolutionary algorithm for hyperparameter tuning in advanced deep learning models. Notably, Fire‐hawk optimization (FHO) emerges as particularly promising, achieving the highest Concordance Index (C‐index) as 0.974 for KIBA datasets and 0.894 for DAVIS datasets and demonstrating exceptional accuracy in ranking continuous predictions across both the datasets.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.0,"publicationDate":"2024-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141820837","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}
Claudia García‐Navarro, Manuel Pulido‐Martos, Cristina Pérez‐Lozano
Engagement has been defined as an attitude toward work, as a positive, satisfying, work‐related state of mind characterized by high levels of vigour, dedication, and absorption. Both its definition and its assessment have been controversial; however, new methods for its assessment, including artificial intelligence (AI), have been introduced in recent years. Therefore, this research aims to determine the state of the art of AI in the study of engagement. To this end, we conducted a systematic review in accordance with PRISMA to analyse the publications to date on the use of AI for the analysis of engagement. The search, carried out in six databases, was filtered, and 15 papers were finally analysed. The results show that AI has been used mainly to assess and predict engagement levels, as well as to understand the relationships between engagement and other variables. The most commonly used AI techniques are machine learning (ML) and natural language processing (NLP), and all publications use structured and unstructured data, mainly from self‐report instruments, social networks, and datasets. The accuracy of the models varies from 22% to 87%, and its main benefit has been to help both managers and HR staff understand employee engagement, although it has also contributed to research. Most of the articles have been published since 2015, and the geography has been global, with publications predominantly in India and the US. In conclusion, this study highlights the state of the art in AI for the study of engagement and concludes that the number of publications is increasing, indicating that this is possibly a new field or area of research in which important advances can be made in the study of engagement through new and novel techniques.
{"title":"The study of engagement at work from the artificial intelligence perspective: A systematic review","authors":"Claudia García‐Navarro, Manuel Pulido‐Martos, Cristina Pérez‐Lozano","doi":"10.1111/exsy.13673","DOIUrl":"https://doi.org/10.1111/exsy.13673","url":null,"abstract":"Engagement has been defined as an attitude toward work, as a positive, satisfying, work‐related state of mind characterized by high levels of vigour, dedication, and absorption. Both its definition and its assessment have been controversial; however, new methods for its assessment, including artificial intelligence (AI), have been introduced in recent years. Therefore, this research aims to determine the state of the art of AI in the study of engagement. To this end, we conducted a systematic review in accordance with PRISMA to analyse the publications to date on the use of AI for the analysis of engagement. The search, carried out in six databases, was filtered, and 15 papers were finally analysed. The results show that AI has been used mainly to assess and predict engagement levels, as well as to understand the relationships between engagement and other variables. The most commonly used AI techniques are machine learning (ML) and natural language processing (NLP), and all publications use structured and unstructured data, mainly from self‐report instruments, social networks, and datasets. The accuracy of the models varies from 22% to 87%, and its main benefit has been to help both managers and HR staff understand employee engagement, although it has also contributed to research. Most of the articles have been published since 2015, and the geography has been global, with publications predominantly in India and the US. In conclusion, this study highlights the state of the art in AI for the study of engagement and concludes that the number of publications is increasing, indicating that this is possibly a new field or area of research in which important advances can be made in the study of engagement through new and novel techniques.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141742192","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}
Federated learning (FL), celebrated for its privacy‐preserving features, has been revealed by recent studies to harbour security vulnerabilities that jeopardize client privacy, particularly through data reconstruction attacks that enable adversaries to recover original client data. This study introduces a client‐level handwriting forgery attack method for FL based on generative adversarial networks (GANs), which reveals security vulnerabilities existing in FL systems. It should be stressed that this research is purely for academic purposes, aiming to raise concerns about privacy protection and data security, and does not encourage illegal activities. Our novel methodology assumes an adversarial scenario wherein adversaries intercept a fraction of parameter updates via victim clients’ wireless communication channels, then use this information to train GAN for data recovery. Finally, the purpose of handwriting imitation is achieved. To rigorously assess and validate our methodology, experiments were conducted using a bespoke Chinese digit dataset, facilitating in‐depth analysis and robust verification of results. Our experimental findings demonstrated enhanced data recovery effectiveness, a client‐level attack and greater versatility compared to prior art. Notably, our method maintained high attack performance even with a streamlined GAN design, yielding increased precision and significantly faster execution times compared to standard methods. Specifically, our experimental numerical results revealed a substantial boost in reconstruction accuracy by 16.7%, coupled with a 51.9% decrease in computational time compared to the latest similar techniques. Furthermore, tests on a simplified version of our GAN exhibited an average 10% enhancement in accuracy, alongside a remarkable 70% reduction in time consumption. By surmounting the limitations of previous work, this study fills crucial gaps and affirms the effectiveness of our approach in achieving high‐accuracy client‐level data reconstruction within the FL context, thereby stimulating further exploration into FL security measures.
联合学习（Federated Learning，FL）因其保护隐私的特性而备受赞誉，但最近的研究却揭示了它存在着危害客户端隐私的安全漏洞，特别是通过数据重构攻击，使对手能够恢复原始客户端数据。本研究介绍了一种基于生成式对抗网络（GANs）的 FL 客户端级笔迹伪造攻击方法，揭示了 FL 系统中存在的安全漏洞。需要强调的是，本研究纯粹出于学术目的，旨在引起人们对隐私保护和数据安全的关注，并不鼓励非法活动。我们的新方法假设了一种对抗场景，即对抗者通过受害者客户端的无线通信信道截获一部分参数更新，然后利用这些信息训练 GAN 进行数据恢复。最后，笔迹模仿的目的就达到了。为了严格评估和验证我们的方法，我们使用定制的中文数字数据集进行了实验，以便对结果进行深入分析和稳健验证。我们的实验结果表明，与现有技术相比，我们的方法提高了数据恢复的有效性、客户端级别的攻击和更大的通用性。值得注意的是，即使采用精简的 GAN 设计，我们的方法仍能保持较高的攻击性能，与标准方法相比，精度更高，执行时间更短。具体来说，我们的数值实验结果表明，与最新的类似技术相比，重建精度大幅提高了 16.7%，计算时间减少了 51.9%。此外，对我们的 GAN 简化版进行的测试表明，精度平均提高了 10%，同时耗时显著减少了 70%。通过克服以往工作的局限性，本研究填补了重要空白，并肯定了我们的方法在 FL 环境下实现高精度客户端级数据重建的有效性，从而激发了对 FL 安全措施的进一步探索。
{"title":"A generative adversarial network‐based client‐level handwriting forgery attack in federated learning scenario","authors":"Lei Shi, Han Wu, Xu Ding, Hao Xu, Sinan Pan","doi":"10.1111/exsy.13676","DOIUrl":"https://doi.org/10.1111/exsy.13676","url":null,"abstract":"Federated learning (FL), celebrated for its privacy‐preserving features, has been revealed by recent studies to harbour security vulnerabilities that jeopardize client privacy, particularly through data reconstruction attacks that enable adversaries to recover original client data. This study introduces a client‐level handwriting forgery attack method for FL based on generative adversarial networks (GANs), which reveals security vulnerabilities existing in FL systems. It should be stressed that this research is purely for academic purposes, aiming to raise concerns about privacy protection and data security, and does not encourage illegal activities. Our novel methodology assumes an adversarial scenario wherein adversaries intercept a fraction of parameter updates via victim clients’ wireless communication channels, then use this information to train GAN for data recovery. Finally, the purpose of handwriting imitation is achieved. To rigorously assess and validate our methodology, experiments were conducted using a bespoke Chinese digit dataset, facilitating in‐depth analysis and robust verification of results. Our experimental findings demonstrated enhanced data recovery effectiveness, a client‐level attack and greater versatility compared to prior art. Notably, our method maintained high attack performance even with a streamlined GAN design, yielding increased precision and significantly faster execution times compared to standard methods. Specifically, our experimental numerical results revealed a substantial boost in reconstruction accuracy by 16.7%, coupled with a 51.9% decrease in computational time compared to the latest similar techniques. Furthermore, tests on a simplified version of our GAN exhibited an average 10% enhancement in accuracy, alongside a remarkable 70% reduction in time consumption. By surmounting the limitations of previous work, this study fills crucial gaps and affirms the effectiveness of our approach in achieving high‐accuracy client‐level data reconstruction within the FL context, thereby stimulating further exploration into FL security measures.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141613269","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}
With the progressive augmentation of parameters in multimodal models, to optimize computational efficiency, some studies have adopted the approach of fine‐tuning the unimodal pre‐training model to achieve multimodal fusion tasks. However, these methods tend to rely solely on simplistic or singular fusion strategies, thereby neglecting more flexible fusion approaches. Moreover, existing methods prioritize the integration of modality features containing highly semantic information, often overlooking the influence of fusing low‐level features on the outcomes. Therefore, this study introduces an innovative approach named multilevel feature fusion guided by prompts (MFF‐GP), a multimodal dynamic fusion framework. It guides the dynamic neural network by prompt vectors to dynamically select the suitable fusion network for each hierarchical feature of the unimodal pre‐training model. This method improves the interactions between multiple modalities and promotes a more efficient fusion of features across them. Extensive experiments on the UPMC Food 101, SNLI‐VE and MM‐IMDB datasets demonstrate that with only a few trainable parameters, MFF‐GP achieves significant accuracy improvements compared to a newly designed PMF based on fine‐tuning—specifically, an accuracy improvement of 2.15% on the UPMC Food 101 dataset and 0.82% on the SNLI‐VE dataset. Further study of the results reveals that increasing the diversity of interactions between distinct modalities is critical and delivers significant performance improvements. Furthermore, for certain multimodal tasks, focusing on the low‐level features is beneficial for modality integration. Our implementation is available at: https://github.com/whq2024/MFF-GP.
{"title":"Multimodal dynamic fusion framework: Multilevel feature fusion guided by prompts","authors":"Lei Pan, Huan‐Qing Wu","doi":"10.1111/exsy.13668","DOIUrl":"https://doi.org/10.1111/exsy.13668","url":null,"abstract":"With the progressive augmentation of parameters in multimodal models, to optimize computational efficiency, some studies have adopted the approach of fine‐tuning the unimodal pre‐training model to achieve multimodal fusion tasks. However, these methods tend to rely solely on simplistic or singular fusion strategies, thereby neglecting more flexible fusion approaches. Moreover, existing methods prioritize the integration of modality features containing highly semantic information, often overlooking the influence of fusing low‐level features on the outcomes. Therefore, this study introduces an innovative approach named multilevel feature fusion guided by prompts (MFF‐GP), a multimodal dynamic fusion framework. It guides the dynamic neural network by prompt vectors to dynamically select the suitable fusion network for each hierarchical feature of the unimodal pre‐training model. This method improves the interactions between multiple modalities and promotes a more efficient fusion of features across them. Extensive experiments on the UPMC Food 101, SNLI‐VE and MM‐IMDB datasets demonstrate that with only a few trainable parameters, MFF‐GP achieves significant accuracy improvements compared to a newly designed PMF based on fine‐tuning—specifically, an accuracy improvement of 2.15% on the UPMC Food 101 dataset and 0.82% on the SNLI‐VE dataset. Further study of the results reveals that increasing the diversity of interactions between distinct modalities is critical and delivers significant performance improvements. Furthermore, for certain multimodal tasks, focusing on the low‐level features is beneficial for modality integration. Our implementation is available at: <jats:ext-link xmlns:xlink=\"http://www.w3.org/1999/xlink\" xlink:href=\"https://github.com/whq2024/MFF-GP\">https://github.com/whq2024/MFF-GP</jats:ext-link>.","PeriodicalId":51053,"journal":{"name":"Expert Systems","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141613581","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：