PeerJ Computer Science最新文献

{"title":"Fake news detection: state-of-the-art review and advances with attention to Arabic language aspects.","authors":"Eman Salamah Albtoush, Keng Hoon Gan, Saif A Ahmad Alrababa","doi":"10.7717/peerj-cs.2693","DOIUrl":"10.7717/peerj-cs.2693","url":null,"abstract":"<p><p>The proliferation of fake news has become a significant threat, influencing individuals, institutions, and societies at large. This issue has been exacerbated by the pervasive integration of social media into daily life, directly shaping opinions, trends, and even the economies of nations. Social media platforms have struggled to mitigate the effects of fake news, relying primarily on traditional methods based on human expertise and knowledge. Consequently, machine learning (ML) and deep learning (DL) techniques now play a critical role in distinguishing fake news, necessitating their extensive deployment to counter the rapid spread of misinformation across all languages, particularly Arabic. Detecting fake news in Arabic presents unique challenges, including complex grammar, diverse dialects, and the scarcity of annotated datasets, along with a lack of research in the field of fake news detection compared to English. This study provides a comprehensive review of fake news, examining its types, domains, characteristics, life cycle, and detection approaches. It further explores recent advancements in research leveraging ML, DL, and transformer-based techniques for fake news detection, with a special attention to Arabic. The research delves into Arabic-specific pre-processing techniques, methodologies tailored for fake news detection in the language, and the datasets employed in these studies. Additionally, it outlines future research directions aimed at developing more effective and robust strategies to address the challenge of fake news detection in Arabic content.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2693"},"PeriodicalIF":3.5,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935763/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712074","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}

{"title":"Cloud-edge MQTT messaging for latency mitigation and broker memory footprint reduction.","authors":"Yi-Hsuan Tseng, Chao Wang, Yu-Tse Wei, Yu-Ting Chiang","doi":"10.7717/peerj-cs.2741","DOIUrl":"10.7717/peerj-cs.2741","url":null,"abstract":"<p><p>The deployment of smart-city applications has increased the number of Internet of Things (IoT) devices connected to a network cloud. Thanks to its flexibility in matching data publishers and subscribers, broker-based data communication could be a solution for such IoT data delivery, and MQTT is one of the widely used messaging protocols in this class. While MQTT by default does not differentiate message flows by size, it is observed that transient local network congestion may cause size-dependent latency additions, and that the accumulation of large message copies in the cloud broker could run out of the broker memory. In response, in the scope of cloud-edge messaging, this research article presents problem analysis, system design and implementation, and empirical and analytical performance evaluation. The article introduces three message scheduling policies for subscribers deployed at network edge, and a memory allocation scheme for MQTT broker deployed at network cloud. The proposed design has been implemented based on Eclipse Mosquitto, an open-source MQTT broker implementation. Empirical and analytical validations have demonstrated the performance of the proposed design in latency mitigation, and the result also shows that, empirically, the proposed design may save the run-time broker memory footprint by about 75%. Applicability of the proposed design to other messaging services are discussed by the end of the article.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2741"},"PeriodicalIF":3.5,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935751/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712128","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}

{"title":"Adversarial feature learning for semantic communication in human 3D reconstruction.","authors":"Shaojiang Liu, Jiajun Zou, Zhendan Liu, Meixia Dong, Zhiping Wan","doi":"10.7717/peerj-cs.2731","DOIUrl":"10.7717/peerj-cs.2731","url":null,"abstract":"<p><p>With the widespread application of human body 3D reconstruction technology across various fields, the demands for data transmission and processing efficiency continue to rise, particularly in scenarios where network bandwidth is limited and low latency is required. This article introduces an Adversarial Feature Learning-based Semantic Communication method (AFLSC) for human body 3D reconstruction, which focuses on extracting and transmitting semantic information crucial for the 3D reconstruction task, thereby significantly optimizing data flow and alleviating bandwidth pressure. At the sender's end, we propose a multitask learning-based feature extraction method to capture the spatial layout, keypoints, posture, and depth information from 2D human images, and design a semantic encoding technique based on adversarial feature learning to encode these feature information into semantic data. We also develop a dynamic compression technique to efficiently transmit this semantic data, greatly enhancing transmission efficiency and reducing latency. At the receiver's end, we design an efficient multi-level semantic feature decoding method to convert semantic data back into key image features. Finally, an improved ViT-diffusion model is employed for 3D reconstruction, producing human body 3D mesh models. Experimental results validate the advantages of our method in terms of data transmission efficiency and reconstruction quality, demonstrating its excellent potential for application in bandwidth-limited environments.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2731"},"PeriodicalIF":3.5,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935756/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712142","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}

{"title":"Data augmentation for Arabic text classification: a review of current methods, challenges and prospective directions.","authors":"Samia F Abdhood, Nazlia Omar, Sabrina Tiun","doi":"10.7717/peerj-cs.2685","DOIUrl":"10.7717/peerj-cs.2685","url":null,"abstract":"<p><p>The effectiveness of data augmentation techniques, <i>i.e</i>., methods for artificially creating new data, has been demonstrated in many domains, from images to textual data. Data augmentation methods were established to manage different issues regarding the scarcity of training datasets or the class imbalance to enhance the performance of classifiers. This review article investigates data augmentation techniques for Arabic texts, specifically in the text classification field. A thorough review was conducted to give a concise and comprehensive understanding of these approaches in the context of Arabic classification. The focus of this article is on Arabic studies published from 2019 to 2024 about data augmentation in Arabic text classification. Inclusion and exclusion criteria were applied to ensure a comprehensive vision of these techniques in Arabic natural language processing (ANLP). It was found that data augmentation research for Arabic text classification dominates sentiment analysis and propaganda detection, with initial studies emerging in 2019; very few studies have investigated other domains like sarcasm detection or text categorization. We also observed the lack of benchmark datasets for performing the tasks. Most studies have focused on short texts, such as Twitter data or reviews, while research on long texts still needs to be explored. Additionally, various data augmentation methods still need to be examined for long texts to determine if techniques effective for short texts are also applicable to longer texts. A rigorous investigation and comparison of the most effective strategies is required due to the unique characteristics of the Arabic language. By doing so, we can better understand the processes involved in Arabic text classification and hence be able to select the most suitable data augmentation methods for specific tasks. This review contributes valuable insights into Arabic NLP and enriches the existing body of knowledge.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2685"},"PeriodicalIF":3.5,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935767/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712168","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}

{"title":"DeepSpoofNet: a framework for securing UAVs against GPS spoofing attacks.","authors":"Aziz Ur Rehman Badar, Danish Mahmood, Adeel Iqbal, Sung Won Kim, Sedat Akleylek, Korhan Cengiz, Ali Nauman","doi":"10.7717/peerj-cs.2714","DOIUrl":"10.7717/peerj-cs.2714","url":null,"abstract":"<p><p>Uncrewed Aerial Vehicles (UAVs) are frequently utilized in several domains such as transportation, distribution, monitoring, and aviation. A significant security vulnerability is the Global Positioning System (GPS) Spoofing attack, wherein the assailant deceives the GPS receiver by transmitting counterfeit signals, thereby gaining control of the UAV. This can result in the UAV being captured or, in certain instances, destroyed. Numerous strategies have been presented to identify counterfeit GPS signals. Although there have been notable advancements in machine learning (ML) for detecting GPS spoofing attacks, there are still challenges and limitations in the current state-of-the-art research. These include imbalanced datasets, sub-optimal feature selection, and the accuracy of attack detection in resource-constrained environments. The proposed framework investigates the optimal pairing of feature selection (FS) methodologies and deep learning techniques for detecting GPS spoofing attacks on UAVs. The primary objective of this study is to address the challenges associated with detecting GPS spoofing attempts in UAVs. The study focuses on tackling the issue of imbalanced datasets by implementing rigorous oversampling techniques. To do this, a comprehensive approach is proposed that combines advanced feature selection techniques with powerful neural network (NN) architectures. The selected attributes from this process are then transmitted to the succeeding tiers of a hybrid NN, which integrates convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) components. The Analysis of Variance (ANOVA) + CNN-BiLSTM hybrid model demonstrates superior performance, producing exceptional results with a precision of 98.84%, accuracy of 99.25%, F1 score of 99.26%, and recall of 99.69%. The proposed hybrid model for detecting GPS spoofing attacks exhibits significant improvements in terms of prediction accuracy, true positive and false positive rates, as well as F1 score and recall values.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2714"},"PeriodicalIF":3.5,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935755/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143711987","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}

{"title":"Customizable pattern synthesis: a deep generative approach for lantern designs.","authors":"Mengran Yan, Chun Tang, Jida Yan, Siti Suhaily Surip","doi":"10.7717/peerj-cs.2732","DOIUrl":"10.7717/peerj-cs.2732","url":null,"abstract":"<p><p>Pattern design is essential in various domains, especially in traditional lantern production, where patterns convey cultural history and artistic values. Our research presents an innovative generative model that produces customizable lantern patterns, integrating classical aesthetics with modern design features <i>via</i> a generative adversarial network (GAN)-based framework. The model was trained on an extensive dataset of over 17,000 pattern images over ten various categories. Experimental assessment demonstrates the model's remarkable proficiency, achieving an Inception Score of 5.259, much surpassing the performance of other GAN-based approaches. This exceptional result demonstrates the effective integration of traditional pattern elements with AI-driven design processes. The model offers enhanced design flexibility <i>via</i> noise vector hybridization and post-processing techniques, allowing for accurate control over pattern production while preserving cultural authenticity. These capabilities make our model a valuable tool for modernizing lantern pattern design while maintaining classic artistic elements.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2732"},"PeriodicalIF":3.5,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935754/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712135","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}

{"title":"A zero-trust based scheme for detecting illegal terminals in the Internet of Things of smart grid.","authors":"Hongyu Zhu, Jianwei Tian, Qian Chen, Zheng Tian, Weiqiang Luo, Mingguang Li","doi":"10.7717/peerj-cs.2736","DOIUrl":"https://doi.org/10.7717/peerj-cs.2736","url":null,"abstract":"<p><p>In recent years, the Internet of Things (IoT) for electricity has faced a series of new challenges. Attackers use a compromised terminal as a springboard to enter the network, steal data, issue malicious commands, and cause great harm. In order to combat the threat of compromised terminals, this article proposes a zero-trust based detection scheme for illegal terminals, based on the principle of \"never trust, always verify\" security mechanism. Firstly, the detection scheme uses the state secret SM9 secret system to authenticate the access device. Then, it proposes a continuous trust evaluation based on the centroid drift trust algorithm on the characteristics of the traffic of the input device. Finally, it generates a real-time access policy by the access control engine to achieve a dynamic access policy. Finally, the access control engine generates real-time access policies to achieve dynamic access control. Experimental results show that the designed system has a high security detection accuracy and can effectively deal with the threat of compromised terminals.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2736"},"PeriodicalIF":3.5,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935753/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712141","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}

{"title":"Comparing diversity, negativity, and stereotypes in Chinese-language AI technologies: an investigation of Baidu, Ernie and Qwen.","authors":"Geng Liu, Carlo Alberto Bono, Francesco Pierri","doi":"10.7717/peerj-cs.2694","DOIUrl":"10.7717/peerj-cs.2694","url":null,"abstract":"<p><p>Large language models (LLMs) and search engines have the potential to perpetuate biases and stereotypes by amplifying existing prejudices in their training data and algorithmic processes, thereby influencing public perception and decision-making. While most work has focused on Western-centric AI technologies, we examine social biases embedded in prominent Chinese-based commercial tools, the main search engine Baidu and two leading LLMs, Ernie and Qwen. Leveraging a dataset of 240 social groups across 13 categories describing Chinese society, we collect over 30 k views encoded in the aforementioned tools by prompting them to generate candidate words describing these groups. We find that language models exhibit a broader range of embedded views compared to the search engine, although Baidu and Qwen generate negative content more often than Ernie. We also observe a moderate prevalence of stereotypes embedded in the language models, many of which potentially promote offensive or derogatory views. Our work highlights the importance of prioritizing fairness and inclusivity in AI technologies from a global perspective.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2694"},"PeriodicalIF":3.5,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935762/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712131","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}

{"title":"Enhanced mutation strategy based differential evolution for global optimization problems.","authors":"Pawan Mishra, Musrrat Ali, Pooja, Safiqul Islam","doi":"10.7717/peerj-cs.2696","DOIUrl":"10.7717/peerj-cs.2696","url":null,"abstract":"<p><p>Differential evolution (DE) stands out as a prominent algorithm for addressing global optimization challenges. The efficacy of DE hinges crucially upon its mutation operation, which serves as a pivotal mechanism in generating diverse and high-quality solutions. This article explores various mutation operations aimed at augmenting the performance of DE in global optimization tasks. A distinct mutation strategy is introduced, with the primary objective of achieving a harmonious equilibrium between exploration and exploitation to enhance both convergence speed and solution quality. The proposed DE centres on a novel mutation-based strategy, introducing a new coefficient factor (\"σ\") in conjunction with the base vector of the basic mutation strategy (\"DE/rand/1\"). This innovation aims to fortify the convergence of local variables during exploitation, thereby improving both the convergence rate and quality. The effectiveness of the proposed mutation operations is evaluated across a set of 27 benchmark functions commonly employed in global optimization. Experimental results conclusively demonstrate that these enhanced mutation strategies significantly outperform state-of-the-art algorithms in terms of solution accuracy and convergence speed. This study underscores the critical role of mutation operations in DE and provides valuable insights for designing more potent mutation strategies to tackle complex global optimization problems.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2696"},"PeriodicalIF":3.5,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935757/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712103","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}

{"title":"A collaborative inference strategy for medical image diagnosis in mobile edge computing environment.","authors":"Shiqian Zhang, Yong Cui, Dandan Xu, Yusong Lin","doi":"10.7717/peerj-cs.2708","DOIUrl":"10.7717/peerj-cs.2708","url":null,"abstract":"<p><p>The popularity and convenience of mobile medical image analysis and diagnosis in mobile edge computing (MEC) environments have greatly improved the efficiency and quality of healthcare services, necessitating the use of deep neural networks (DNNs) for image analysis. However, DNNs face performance and energy constraints when operating on the mobile side, and are limited by communication costs and privacy issues when operating on the edge side, and previous edge-end collaborative approaches have shown unstable performance and low search efficiency when exploring classification strategies. To address these issues, we propose a DNN edge-optimized collaborative inference strategy (MOCI) for medical image diagnosis, which optimizes data transfer and computation allocation by combining compression techniques and multi-agent reinforcement learning (MARL) methods. The MOCI strategy first uses coding and quantization-based compression methods to reduce the redundancy of image data during transmission at the edge, and then dynamically segments the DNN model through MARL and executes it collaboratively between the edge and the mobile device. To improve policy stability and adaptability, MOCI introduces the optimal transmission distance (Wasserstein) to optimize the policy update process, and uses the long short-term memory (LSTM) network to improve the model's adaptability to dynamic task complexity. The experimental results show that the MOCI strategy can effectively solve the collaborative inference task of medical image diagnosis and significantly reduce the latency and energy consumption with less than a 2% loss in classification accuracy, with a maximum reduction of 38.5% in processing latency and 71% in energy consumption compared to other inference strategies. In real-world MEC scenarios, MOCI has a wide range of potential applications that can effectively promote the development and application of intelligent healthcare.</p>","PeriodicalId":54224,"journal":{"name":"PeerJ Computer Science","volume":"11 ","pages":"e2708"},"PeriodicalIF":3.5,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11935760/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143712120","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}