WIREs Data Mining and Knowledge Discovery最新文献

{"title":"A Guide to Machine Learning Epistemic Ignorance, Hidden Paradoxes, and Other Tensions","authors":"M. Z. Naser","doi":"10.1002/widm.70038","DOIUrl":"https://doi.org/10.1002/widm.70038","url":null,"abstract":"Machine learning (ML) has rapidly scaled in capacity and complexity, yet blind spots persist beneath its high performance façade. In order to shed more light on this argument, this paper presents a curated catalogue of 175 unconventional concepts, each capturing a paradox, tension, or overlooked risk in modern ML practice. Through nine themes spanning data quality, model architecture and training, interpretability and explainability, fairness and bias, model behavior and limitations, evaluation and metrics, multimodal and system integration, practical and societal implications, and causal reasoning, we provide conceptual definitions, illustrative examples, and actionable mitigation strategies. This review equips practitioners and researchers with a structured taxonomy for diagnosing and preempting the brittle edges of modern ML systems and offers a paradox detection and remediation framework (PDRF) to anticipate limitations, design more thoughtful evaluation protocols, and develop ML systems that balance predictive power with epistemic transparency.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Fundamental Concepts of Data and Knowledge &gt; Data Concepts</jats:list-item> <jats:list-item>Fundamental Concepts of Data and Knowledge &gt; Big Data Mining</jats:list-item> <jats:list-item>Technologies &gt; Computational Intelligence</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144693602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Statistical and Machine Learning Approaches for Electrical Energy Forecasting","authors":"Solange Machado, Xingquan Zhu","doi":"10.1002/widm.70033","DOIUrl":"https://doi.org/10.1002/widm.70033","url":null,"abstract":"With renewable energy being aggressively integrated into the grid, energy supplies are becoming vulnerable to weather and the environment, and are often incapable of meeting population demands at a large scale if not accurately predicted for energy planning. Understanding consumers' power demands ahead of time and the influences of weather on consumption and generation can help producers generate effective power management plans to support the target demand. In addition to the high correlation with the environment, consumers' behaviors also cause non‐stationary characteristics of energy data, which is the main challenge for energy prediction. In this survey, we perform a review of the literature on prediction methods in the energy field. So far, most of the available research encompasses one type of generation or consumption. There is no research approaching prediction in the energy sector as a whole and its correlated features. We propose to address the energy prediction challenges from both consumption and generation sides, encompassing techniques from statistical to machine learning techniques. We also summarize the work related to energy prediction, electricity measurements, challenges related to energy consumption and generation, energy forecasting methods, and real‐world energy forecasting resources, such as datasets and software solutions for energy prediction.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Application Areas &gt; Industry Specific Applications</jats:list-item> <jats:list-item>Technologies &gt; Prediction</jats:list-item> <jats:list-item>Technologies &gt; Machine Learning</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"671 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144629782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Systematic Literature Survey of Crowdsourcing: Current Status and Future Perspectives","authors":"Himanshu Suyal, Avtar Singh","doi":"10.1002/widm.70037","DOIUrl":"https://doi.org/10.1002/widm.70037","url":null,"abstract":"Crowdsourcing has recently evolved as a distributed human problem‐solving method and has received considerable interest from academics and practitioners in various domains. The proliferation of crowdsourcing has made it much simpler to utilize the intelligence and adaptability of many people to learn new knowledge to solve the problem of acquiring new knowledge. In the past, numerous crowdsourcing works have highlighted multiple aspects; however, no surveys have been conducted that focus on the entire crowdsourcing process. This concentrated survey provides a comprehensive review of the technical advances from a systematic perspective. This survey systematically reviews technical advances for a crowdsourcing process that contains four dimensions: task modeling, crowdsourcing data acquisition, the learning process, and predictive model learning, and proposes a comprehensive and scalable framework from CROWD4AI (Crowdsourcing Framework with 4 Dimensions for Artificial Intelligence). In addition, this paper focuses on each dimension's potential challenges and future direction, encouraging researchers to participate in crowdsourcing. To bridge theory with practice, we also include a detailed case study that demonstrates the real‐world application of our proposed framework in the context of annotating cultural heritage damages using crowdsourced input. The case study illustrates how the framework supports effective task design, label collection, robust learning strategies, and accurate predictive modeling in a practical setting.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Technologies &gt; Crowdsourcing</jats:list-item> <jats:list-item>Technologies &gt; Machine Learning</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"52 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144629784","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine Learning and Deep Learning Techniques to Detect Mental Stress Using Various Physiological Signals: A Critical Insight","authors":"Megha Khandelwal, Arun Sharma","doi":"10.1002/widm.70035","DOIUrl":"https://doi.org/10.1002/widm.70035","url":null,"abstract":"This paper presents a comprehensive review on the various techniques and methodologies employed to detect stress among individuals. The review encompasses a broad spectrum of methods, including physiological measurements, wearable technology, machine learning and deep learning algorithms, and contactless image‐based techniques. The paper outlines the physiological markers commonly associated with stress, such as Electrocardiogram (ECG), Electroencephalography (EEG), Photoplethysmography (PPG), and Skin Galvanic response. It examines the various wearable and contactless techniques to acquire data. Furthermore, it explores the integration of machine learning and deep learning techniques for the development of predictive stress detection models, highlighting their accuracy. It also addresses the potential of multispectral and hyperspectral imaging in this area. Some of the publicly available datasets are also discussed in this paper.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Application Areas &gt; Health Care</jats:list-item> <jats:list-item>Technologies &gt; Machine Learning</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144629783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Survey on Efficient Vision‐Language Models","authors":"Gaurav Shinde, Anuradha Ravi, Emon Dey, Shadman Sakib, Milind Rampure, Nirmalya Roy","doi":"10.1002/widm.70036","DOIUrl":"https://doi.org/10.1002/widm.70036","url":null,"abstract":"Vision‐language models (VLMs) integrate visual and textual information, enabling a wide range of applications such as image captioning and visual question answering, making them crucial for modern AI systems. However, their high computational demands pose challenges for real‐time applications. This has led to a growing focus on developing efficient vision‐language models. In this survey, we review key techniques for optimizing VLMs on edge and resource‐constrained devices. We also explore compact VLM architectures, frameworks, and provide detailed insights into the performance–memory trade‐offs of efficient VLMs. Furthermore, we establish a GitHub repository at MPSC‐GitHub to compile all surveyed papers, which we will actively update. Our objective is to foster deeper research in this area.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Fundamental Concepts of Data and Knowledge &gt; Big Data Mining</jats:list-item> <jats:list-item>Technologies &gt; Internet of Things</jats:list-item> <jats:list-item>Technologies &gt; Artificial Intelligence</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"23 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144622234","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of Explainable Artificial Intelligence (XAI) Techniques in Patients With Intracranial Hemorrhage: A Systematic Review","authors":"Ali Kohan, Amir Zahedi, Roohallah Alizadehsani, Ru‐San Tan, U. Rajendra Acharya","doi":"10.1002/widm.70031","DOIUrl":"https://doi.org/10.1002/widm.70031","url":null,"abstract":"Intracranial hemorrhage (IH) is a critical condition requiring rapid and accurate diagnosis to ensure effective treatment and reduce mortality rates. Recently, artificial intelligence (AI) models have demonstrated significant potential in automating the detection and analysis of brain injuries in IH patients. However, the “black‐box” nature of many AI systems raises concerns about transparency, reliability, and clinical applicability. Explainable AI (XAI) addresses these challenges by making AI models more interpretable, allowing healthcare professionals to understand and trust the decision‐making processes. This review paper explores various XAI techniques—such as SHapley Additive exPlanations (SHAP), Local Interpretable Model‐Agnostic Explanations (LIME), Randomized Input Sampling for Explanation (RISE), Class Activation Mapping (CAM), and its variants—and their specific applications in IH clinical tasks. We systematically examine studies incorporating XAI for curing IH patients, highlighting how these methods enhance model transparency and support clinical decision‐making. The Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) methodology was employed to select the papers. Studies are categorized into those using tabular data and those using image data. The literature indicates a rapidly growing number of XAI publications in this field. SHAP is the most commonly used XAI method for tabular data, while CAM‐based methods, such as Grad‐CAM, dominate in image‐based applications. Furthermore, we discuss current limitations of XAI methods and future research directions. This review aims to provide researchers and clinicians with valuable insights into the role of XAI in improving the reliability and practical integration of AI‐driven tools for IH patient care.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Application Areas &gt; Health Care</jats:list-item> <jats:list-item>Fundamental Concepts of Data and Knowledge &gt; Explainable AI</jats:list-item> <jats:list-item>Technologies &gt; Machine Learning</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"48 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144503560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence Techniques Enabled Soil Moisture Estimation Frameworks Using Remote Sensing Satellite Images: Challenges and Future Directions‐ Review","authors":"Mangayarkarasi Ramaiah, Prabhavathy Settu, Vinayakumar Ravi","doi":"10.1002/widm.70032","DOIUrl":"https://doi.org/10.1002/widm.70032","url":null,"abstract":"Forecasting soil moisture is critical for keeping groundwater levels stable, monitoring droughts, and assisting agricultural productivity. Surface soil moisture has a tremendous impact on both the environment and society. To provide proper soil moisture, the right tools are required. Gravimetric, physical, and empirical models produce reliable results, but they are generally context‐dependent and inappropriate for large‐scale investigations. Remote sensing has developed as a credible technology for estimating large‐scale soil moisture levels. However, various obstacles exist when getting soil moisture data using remote sensing, including the availability and precision of data sources. The spatial and temporal limits of many remote sensing sources, such as microwave and optical sensors, combined with environmental conditions, provide considerable feasibility issues. As a result, a robust model capable of accurately capturing both linear and nonlinear connections between multiple surface soil variables is critical. Recently, AI approaches have been identified as promising options for managing complicated factors in this domain. This review paper investigates the use of several AI algorithms for estimating soil moisture content (SMC). It focusses on AI‐enabled frameworks built with remote sensing satellite imagery. In addition to including in situ observations, the study discusses the advantages of AI approaches, the issues they solve, and provides a detailed description of the integration of microwave, optical, and combination (synergistic) data sources. This paper also addresses the most common AI approaches applied with various types of remote sensing data and the results they produced. By exploring the strengths and technical problems associated with diverse data sources, this work hopes to help researchers make wise choices about data selection and model construction. Finally, the proposed future research directions are likely to assist emerging researchers in broadening the scope of this critical topic in a way that corresponds with future demands.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Technologies &gt; Artificial Intelligence</jats:list-item> <jats:list-item>Technologies &gt; Machine Learning</jats:list-item> <jats:list-item>Technologies &gt; Prediction</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"16 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144503606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Literature Review of Textual Cyber Abuse Detection Using Cutting‐Edge Natural Language Processing Techniques: Language Models and Large Language Models","authors":"J. Angel Diaz‐Garcia, Joao Paulo Carvalho","doi":"10.1002/widm.70029","DOIUrl":"https://doi.org/10.1002/widm.70029","url":null,"abstract":"The success of social media platforms has facilitated the emergence of various forms of online abuse within digital communities. This abuse manifests in multiple ways, including hate speech, cyberbullying, emotional abuse, grooming, and shame sexting or sextortion. In this paper, we present a comprehensive analysis of the different forms of abuse prevalent in social media, with a particular focus on how emerging technologies, such as Language Models (LMs) and Large Language Models (LLMs), are reshaping both the detection and generation of abusive content within these networks. We delve into the mechanisms through which social media abuse is perpetuated, exploring the psychological and social impact. To achieve this, we conducted a literature review based on PRISMA methodology, deriving key insights in the field of cyber abuse detection. Additionally, we examine the dual role of advanced language models—highlighting their potential to enhance automated detection systems for abusive behavior while also acknowledging their capacity to generate harmful content. This paper contributes to the ongoing discourse on online safety and ethics by offering both theoretical and practical insights into the evolving landscape of cyber abuse, as well as the technological innovations that simultaneously mitigate and exacerbate it. The findings support platform administrators and policymakers in developing more effective moderation strategies, conducting comprehensive risk assessments, and integrating AI responsibly to create safer digital environments.This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item>Algorithmic Development &gt; Web Mining</jats:list-item> <jats:list-item>Technologies &gt; Classification</jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"67 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144503457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Overview of Heterogeneous Social Network Analysis","authors":"Deepti Singh, Ankita Verma","doi":"10.1002/widm.70028","DOIUrl":"https://doi.org/10.1002/widm.70028","url":null,"abstract":"Heterogeneous Social Networks (HSNs) represent complex structures where diverse entities, such as users, items, and interactions, coexist and interact within a unified framework. This paper offers a systematic review of HSN Analysis, addressing the theoretical and practical challenges associated with investigating the interplay between varied node types and diverse relationships within HSNs. The paper begins by defining HSNs and outlining their characteristics, highlighting the existence of diverse entity kinds and a range of relationship types. It explores the significance of HSNs in modeling real‐world systems, including online social platforms, biological networks, e‐commerce networks, and recommendation systems, where diverse entities play distinct roles. The analysis of HSNs extends beyond traditional homogeneous networks, incorporating various types of nodes and edges, and introduces novel considerations for effective analysis. The difficulties in modeling, representing, and analyzing HSNs will be covered in this work. Several reviews of social network analysis have been published in the past, but they often focus on simple networks, not HSN analysis specifically. This paper aims to fill that gap by comprehensively reviewing different aspects of HSN and its analysis. We start with the fundamentals of HSNs, explore its major types‐multi‐relational networks and multi‐modal networks and further their impact on popular data mining tasks. Then, we explore various applications of heterogeneous information network analysis, like recommender systems, text mining, fraud detection, and e‐commerce. Finally, we look at recent research and suggest promising future directions in the field of HSN analysis.","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"44 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144288333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Vehicle Damage Detection Using Artificial Intelligence: A Systematic Literature Review","authors":"Md Jahid Hasan, Cong Kha Nguyen, Yee Ling Boo, Hamed Jahani, Kok-Leong Ong","doi":"10.1002/widm.70027","DOIUrl":"https://doi.org/10.1002/widm.70027","url":null,"abstract":"Automating vehicle damage detection is essential for automotive industry applications like insurance claims, online sales, and repair cost estimates, addressing the labor-intensive, time-consuming, and error-prone nature of current manual inspections. This systematic literature review explores the use of artificial intelligence (AI), particularly deep learning-based algorithms, to improve the accuracy and efficiency of damage detection under dynamic and challenging conditions specific to the requirements of our industry partners. The review is structured around five key research questions and includes extensive empirical evaluations to identify gaps and challenges in existing methods. Findings reveal significant potential for AI to automate and enhance the damage detection process but also highlight areas requiring further research and development. The review discusses these gaps in detail, providing a comprehensive foundation for future work in this field. Furthermore, the review findings are intended to guide both our research and the broader research community in advancing the practical application of AI for vehicle damage assessment. The insights gained from this review are crucial for developing robust AI solutions that can operate effectively in real-world scenarios, ultimately improving operational efficiency and customer experience in the automotive industry.","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"70 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144237453","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}