Computer Science Review最新文献

{"title":"What can artificial intelligence do for soil health in agriculture?","authors":"Stefan Schweng ,&nbsp;Luca Bernardini ,&nbsp;Katharina Keiblinger ,&nbsp;Hans-Peter Kaul ,&nbsp;Iztok Fister Jr. ,&nbsp;Niko Lukač ,&nbsp;Javier Del Ser ,&nbsp;Andreas Holzinger","doi":"10.1016/j.cosrev.2025.100832","DOIUrl":"10.1016/j.cosrev.2025.100832","url":null,"abstract":"<div><div>The integration of artificial intelligence (AI) into soil research presents significant opportunities to advance the understanding, management, and conservation of soil ecosystems. This paper reviews the diverse applications of AI in soil health assessment, predictive modeling of soil properties, and the development of pedotransfer functions within the context of agriculture, emphasizing AI’s advantages over traditional analytical methods. We identify soil organic matter decline, compaction, and biodiversity loss as the most frequently addressed forms of soil degradation. Strong trends include the creation of digital soil maps, particularly for soil organic carbon and chemical properties using remote sensing or easily measurable proxies, as well as the development of decision support systems for crop rotation planning and IoT-based monitoring of soil health and crop performance. While random forest models dominate, support vector machines and neural networks are also widely applied for soil parameter modeling. Our analysis of datasets reveals clear regional biases, with tropical, arid, mild continental, and polar tundra climates remaining underrepresented despite their agricultural relevance. We also highlight gaps in predictor–response combinations for soil property modeling, pointing to promising research avenues such as estimating heavy metal content from soil mineral nitrogen content, microbial biomass, or earthworm abundance. Finally, we provide practical guidelines on data preparation, feature extraction, and model selection. Overall, this study synthesizes recent advances, identifies methodological limitations, and outlines a roadmap for future research, underscoring AI’s transformative potential in soil science.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100832"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145181248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive review of recommender systems: Transitioning from theory to practice","authors":"Shaina Raza ,&nbsp;Mizanur Rahman ,&nbsp;Safiullah Kamawal ,&nbsp;Armin Toroghi ,&nbsp;Ananya Raval ,&nbsp;Farshad Navah ,&nbsp;Amirmohammad Kazemeini","doi":"10.1016/j.cosrev.2025.100849","DOIUrl":"10.1016/j.cosrev.2025.100849","url":null,"abstract":"<div><div>Recommender Systems (RS) play an integral role in enhancing user experiences by providing personalized item suggestions. This survey reviews the progress in RS inclusively from 2017 to 2024, effectively connecting theoretical advances with practical applications. We explore the development from traditional RS techniques like content-based and collaborative filtering to advanced methods involving deep learning, graph-based models, reinforcement learning, and large language models. We also discuss specialized systems such as context-aware, review-based, and fairness-aware RS. The primary goal of this survey is to bridge theory with practice. It addresses challenges across various sectors, including e-commerce, healthcare, and finance, emphasizing the need for scalable, real-time, and trustworthy solutions. Through this survey, we promote stronger partnerships between academic research and industry practices. The insights offered by this survey aim to guide industry professionals in optimizing RS deployment and to inspire future research directions, especially in addressing emerging technological and societal trends. The survey resources are available in the public GitHub repository <span><span>https://github.com/VectorInstitute/Recommender-Systems-Survey</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100849"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145499403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Parameterized algorithms on geometric intersection graphs","authors":"Jie Xue ,&nbsp;Meirav Zehavi","doi":"10.1016/j.cosrev.2025.100796","DOIUrl":"10.1016/j.cosrev.2025.100796","url":null,"abstract":"<div><div>We survey results on the parameterized complexity of various problems on geometric intersection graphs, particularly (unit) disk graphs. Specifically, we consider: (i) vertex-deletion, packing and pattern detection problems; (ii) <span>Independent Set</span> and <span>Dominating Set</span>; (iii) cut and connectivity problems; (iv) obstacle-removal problems. The discussions include introductions of some of the proof ideas and directions for future research.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100796"},"PeriodicalIF":12.7,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144779484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A survey on decentralized identity management systems","authors":"Jia Xian,&nbsp;Lin You,&nbsp;Qifei Yi,&nbsp;Jixiang Wang,&nbsp;Gengran Hu","doi":"10.1016/j.cosrev.2025.100811","DOIUrl":"10.1016/j.cosrev.2025.100811","url":null,"abstract":"<div><div>Digital identity is a crucial tool in the field such as healthcare, education and government. It serves as the foundation for establishing presence in cyberspace. With the increasing demands for security, control and portability, digital identity has evolved through four stages: centralized, federated, user-centric and self-sovereign identity. Among these, the decentralized identity represents the self-sovereign identity stage, emphasizing the idea of the user fully control over the user’s identity. Blockchain, as the key technology enabling decentralization, provides robust support for the decentralized identity through its transparent, distributed and immutable characteristics. We investigate the blockchain-based Decentralized Identity Management Systems (DIDMS), analyzing their advantages over the previous systems in terms of the enhanced control, privacy, and security. We propose a four-phase architecture for the DIDMS, detailing how the users can achieve self-management through the DIDMS without relying on the centralized entities. Furthermore, this investigation compares the various solutions with the core practical properties, explores the relationships between these features, and summarizes the evaluation criteria for the DIDMS. Additionally, the work evaluates both classical and cutting-edge DIDMS solutions and discusses the challenges faced by the DIDMS, along with potential solutions, providing direction for future research and practice.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100811"},"PeriodicalIF":12.7,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144922624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From past to present: A survey of malicious URL detection techniques, datasets and code repositories","authors":"Ye Tian ,&nbsp;Yanqiu Yu ,&nbsp;Jianguo Sun ,&nbsp;Yanbin Wang","doi":"10.1016/j.cosrev.2025.100810","DOIUrl":"10.1016/j.cosrev.2025.100810","url":null,"abstract":"<div><div>Malicious URLs persistently threaten the cybersecurity ecosystem, by either deceiving users into divulging private data or distributing harmful payloads to infiltrate host systems. The detection of malicious URLs is a protracted arms race between defenders and attackers. Gaining timely insights into the current state of this ongoing battle holds significant importance. However, existing reviews suffer from four critical limitations: 1) Their reliance on algorithm-centric taxonomies obscures understanding of how detection approaches exploit specific modal information channels; 2) They fail to incorporate pivotal LLM/Transformer-based defenses; 3) No open-source implementations are collected to facilitate benchmarking; 4) Insufficient dataset coverage.</div><div>This paper presents a comprehensive review of malicious URL detection technologies, systematically analyzing methods from traditional blacklisting to advanced deep learning approaches (e.g., Transformer, GNNs, and LLMs). Unlike prior surveys, we propose a novel modality-based taxonomy that categorizes existing works according to their primary data modalities (e.g., lexical URL features, HTML structure, JavaScript behavior, visual layout).For instance, we group models that parse DOM trees and extract HTML tag paths into the HTML modality, while those using rendered webpage screenshots are classified under the visual modality. This taxonomy reveals how distinct input channels inform model design, offering new perspectives that are obscured in algorithm-based classifications. This hierarchical classification enables both rigorous technical analysis and clear understanding of multimodal information utilization. Furthermore, to establish a profile of accessible datasets and address the lack of standardized benchmarking (where current studies often lack proper baseline comparisons), we curate and analyze two key resources: 1) publicly available datasets (2016–2024), and 2) open-source implementations from published works (2013–2025). To facilitate cross-method comparison and support future benchmarking efforts, we compile a comparative table summarizing key performance metrics (e.g., Accuracy, F1 Score, AUC) as reported in the original works or open-source repositories. While no formal evaluation protocol is proposed, this effort provides a practical reference point and highlights the need for more standardized benchmarking practices in malicious URL detection research. The review concludes by examining emerging challenges and proposing actionable directions for future research. We maintain a GitHub repository for ongoing curation of datasets and open-source implementations: <span><span>https://github.com/sevenolu7/Malicious-URL-Detection-Open-Source/tree/master</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100810"},"PeriodicalIF":12.7,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144902472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"NLP-based techniques for Cyber Threat Intelligence","authors":"Marco Arazzi ,&nbsp;Dincy R. Arikkat ,&nbsp;Serena Nicolazzo ,&nbsp;Antonino Nocera ,&nbsp;Rafidha Rehiman K.A. ,&nbsp;Vinod P. ,&nbsp;Mauro Conti","doi":"10.1016/j.cosrev.2025.100765","DOIUrl":"10.1016/j.cosrev.2025.100765","url":null,"abstract":"<div><div>In the digital era, threat actors employ sophisticated techniques for which, often, digital traces in the form of textual data are available. Cyber Threat Intelligence (CTI) is related to all the solutions inherent to data collection, processing, and analysis useful to understand a threat actor’s targets and attack behavior. Currently, CTI is assuming an always more crucial role in identifying and mitigating threats and enabling proactive defense strategies. In this context, NLP, an artificial intelligence branch, has emerged as a powerful tool for enhancing threat intelligence capabilities. This survey paper provides a comprehensive overview of NLP-based techniques applied in the context of threat intelligence. It begins by describing the foundational definitions and principles of CTI as a major tool for safeguarding digital assets. It then undertakes a thorough examination of NLP-based techniques for CTI data crawling from Web sources, CTI data analysis, Relation Extraction from cybersecurity data, CTI sharing and collaboration, security threats of CTI, and role of LLM in this domain. Finally, the challenges and limitations of NLP in threat intelligence are exhaustively examined, including data quality issues and ethical considerations. This survey draws a complete framework and serves as a valuable resource for security professionals and researchers seeking to understand the state-of-the-art NLP-based threat intelligence techniques and their potential impact on cybersecurity.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100765"},"PeriodicalIF":13.3,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive review on machine learning-based VPN detection: Scenarios, methods, and open challenges","authors":"Alejandro Guerra-Manzanares ,&nbsp;Maurantonio Caprolu ,&nbsp;Roberto Di Pietro","doi":"10.1016/j.cosrev.2025.100781","DOIUrl":"10.1016/j.cosrev.2025.100781","url":null,"abstract":"<div><div>Virtual Private Networks (VPNs) are an essential tool to protect user privacy and enforce secure communications over the Internet. However, they can also be misused to bypass legit network security mechanisms and hence access otherwise restricted content. These reasons, combined with the fact that VPN supporting technology has continuously evolved—reaching quite a relevant level of sophistication—make detecting VPN traffic a vested research issue for both academia and industry. In this paper, we provide a comprehensive review of machine learning-based (ML) solutions for VPN traffic detection. In particular, we start with framing the problem and identifying the main scenarios and related adversary models. Then, we provide a thorough analysis of the related literature and state-of-the-art in ML methodologies for VPN detection, identifying research gaps and unresolved challenges. In particular, we show that the vast majority of the current solutions rely on a specific dataset that suffers from a few severe limitations, hence questioning the validity of reported results when applied to real use case scenarios. Finally, we summarize existing knowledge highlighting common mistakes and providing guidelines as well as future research directions. To the best of our knowledge, this is the first paper that provides a deep dive into ML methodologies for VPN detection, showing current pitfalls, providing actionable recommendations, as well as suggesting research directions.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100781"},"PeriodicalIF":13.3,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144654535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive survey on sentiment analysis: Framework, techniques, and applications","authors":"Manish Kumar Chandan,&nbsp;Shrabanti Mandal","doi":"10.1016/j.cosrev.2025.100777","DOIUrl":"10.1016/j.cosrev.2025.100777","url":null,"abstract":"<div><div>The study of sentiment analysis (SA), also recognized as opinion mining, is a rapidly emerging area of study in natural language processing (NLP). This area focuses on identifying and extracting emotions and opinions from textual data, categorizing them as either positive, neutral, or negative. Nowadays, most of the people express their opinions on social networking platforms, often using their native languages. The rapid growth of Internet-based applications has given rise to a vast amount of personalized information and broad array of user reviews available online. There are a substantial number of pertinent reviews about a particular domain, which remain difficult for humans to process. Therefore, analyzing user opinions is crucial to extract meaningful insights and understand sentiments effectively. This survey comprehensively examines the spectrum of applications for sentiment analysis within the context of current studies. We then critically review experimental outcomes and limitations observed in cutting-edge studies. Furthermore, we explore lexicon-based methods, machine learning (ML), and deep learning (DL) strategies, as well as emerging techniques like transfer learning, large language models, and multimodal approaches, discussing their strengths as well as their weaknesses.</div><div>In addition, we employed multiple ML and DL strategies, leveraging the two benchmark IMDb and Yelp datasets. Following this, we utilized a systematic framework that incorporated preprocessing techniques, feature extraction, and evaluation metrics to facilitate comprehensive understanding and ensure model generalization. Finally, this study bridges the gap between traditional methods and modern innovations, addressing various challenges in sentiment analysis and proposing a roadmap for future research to mitigate these issues. This article serves as a guiding resource for researchers aiming to build an effective sentiment analysis framework.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100777"},"PeriodicalIF":13.3,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144272311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Chaos-based video encryption techniques: A review","authors":"Suo Gao ,&nbsp;Rui Wu ,&nbsp;Herbert Ho-Ching Iu ,&nbsp;Ugur Erkan ,&nbsp;Yinghong Cao ,&nbsp;Qi Li ,&nbsp;Abdurrahim Toktas ,&nbsp;Jun Mou","doi":"10.1016/j.cosrev.2025.100816","DOIUrl":"10.1016/j.cosrev.2025.100816","url":null,"abstract":"<div><div>Chaos theory, as a powerful tool for studying the nonlinear behavior of dynamic systems, has shown great potential in the field of cryptography. Leveraging the high sensitivity of chaotic systems to initial conditions and their complex dynamic characteristics, chaos-based encryption methods offer enhanced security and unpredictability in the processes of data encryption and decryption. Amid the growing demand for information security, traditional encryption methods face significant challenges, particularly in terms of resistance to attacks and the randomness of key generation. The integration of chaos theory provides new solutions to these challenges. This paper provides a comprehensive review of chaos-based video encryption techniques. First, a mathematical model for video encryption is established, and the strengths and limitations of conventional encryption algorithms when applied to video data are analyzed. Next, we explored the unique role of the complexity and randomness of chaotic systems in encryption design and their impact on enhancing video transmission and storage security. The study then classifies the most common chaos systems and conducts an integrated evaluation of their performance metrics, offering theoretical guidance for selecting suitable chaos systems in video encryption. Building on this foundation, existing chaos-based video encryption algorithms are systematically categorized and reviewed, with their design methodologies and evaluation criteria summarized. Finally, the paper outlines future research directions for chaos-based video encryption and discusses its application potential in intelligent surveillance, medical imaging, military communications, and other fields.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100816"},"PeriodicalIF":12.7,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144931667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A decade systematic review of fusion techniques in financial market prediction","authors":"Soheila Mehrmolaei,&nbsp;Mohammad Saniee Abadeh","doi":"10.1016/j.cosrev.2025.100813","DOIUrl":"10.1016/j.cosrev.2025.100813","url":null,"abstract":"<div><div>Financial markets are structured systems that facilitate investment, trading, and economic growth at both global and local levels. Accurate and reliable prediction in these markets is essential for making profitable decisions. However, the inherent complexity of financial markets, influenced by numerous external factors, makes it difficult to perform analysis and forecasting. In this context, fusion techniques have emerged as a strong and well-known approach, integrating data and features from multiple sources to enhance prediction accuracy. This paper presents a systematic review of studies on fusion techniques in financial market prediction, covering research published between 2016 and 2025. The primary objective is to provide a comprehensive understanding of the role of fusion techniques in financial market prediction from both macro and micro perspectives. To achieve this, we categorize fusion techniques based on level of integration, analyzing their benefits, limitations, and applications. Additionally, we discuss the necessity of fusion approaches, open challenges, and potential future advancements in this domain. Our review emphasizes the growing adoption of multimodal text data fusion using large language models (LLMs) as a promising trend to enhance prediction reliability. Also, we identify key research directions and emerging trends that are expected to shape the future development of fusion-based financial market prediction methods.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"58 ","pages":"Article 100813"},"PeriodicalIF":12.7,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144912513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}