Software refactoring focuses on improving software quality by applying changes to the internal structure that do not alter the observable behavior. Determining which refactorings should be applied and presented to developers the most relevant and optimal refactorings is often challenging. Existing literature suggests that one of the potential sources to identify and recommend required refactorings is the past software development and evolution histories which are often archived in software repositories. In this article, we review a selection of existing literature that has attempted to propose approaches that facilitate refactoring by exploiting information mined from software repositories. Based on the reviewed papers, existing works leverage software history mining to support analysis of code smells, refactoring, and guiding software changes. First, past history information is used to detect design flaws in source code commonly referred to as code smells. Moreover, other studies analyze the evolution of code smells to establish how and when they are introduced into the code base and get resolved. Second, software repositories mining provides useful insights that can be used in predicting the need for refactoring and what specific refactoring operations are required. In addition, past history can be used in detecting and analyzing previously applied refactorings to establish software change facts, for instance, how developers refactor code and the motivation behind it. Finally, change patterns are used to predict further changes that might be required and recommend a set of files for change during a given modification task. The paper further suggests other exciting possibilities that can be pursued in the future in this research direction.
{"title":"Research on mining software repositories to facilitate refactoring","authors":"Ally S. Nyamawe","doi":"10.1002/widm.1508","DOIUrl":"https://doi.org/10.1002/widm.1508","url":null,"abstract":"Software refactoring focuses on improving software quality by applying changes to the internal structure that do not alter the observable behavior. Determining which refactorings should be applied and presented to developers the most relevant and optimal refactorings is often challenging. Existing literature suggests that one of the potential sources to identify and recommend required refactorings is the past software development and evolution histories which are often archived in software repositories. In this article, we review a selection of existing literature that has attempted to propose approaches that facilitate refactoring by exploiting information mined from software repositories. Based on the reviewed papers, existing works leverage software history mining to support analysis of code smells, refactoring, and guiding software changes. First, past history information is used to detect design flaws in source code commonly referred to as code smells. Moreover, other studies analyze the evolution of code smells to establish how and when they are introduced into the code base and get resolved. Second, software repositories mining provides useful insights that can be used in predicting the need for refactoring and what specific refactoring operations are required. In addition, past history can be used in detecting and analyzing previously applied refactorings to establish software change facts, for instance, how developers refactor code and the motivation behind it. Finally, change patterns are used to predict further changes that might be required and recommend a set of files for change during a given modification task. The paper further suggests other exciting possibilities that can be pursued in the future in this research direction.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90643900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
R. Khan, Janani Surya, Maitreyee Roy, M. S. Swathi Priya, Sashwanthi Mohan, S. Raman, Akshay Raman, Abhishek Vyas, R. Raman
The rise of non‐invasive, rapid, and widely accessible quantitative high‐resolution imaging methods, such as modern retinal photography and optical coherence tomography (OCT), has significantly impacted ophthalmology. These techniques offer remarkable accuracy and resolution in assessing ocular diseases and are increasingly recognized for their potential in identifying ocular biomarkers of systemic diseases. The application of artificial intelligence (AI) has been demonstrated to have promising results in identifying age, gender, systolic blood pressure, smoking status, and assessing cardiovascular disorders from the fundus and OCT images. Although our understanding of eye–body relationships has advanced from decades of conventional statistical modeling in large population‐based studies incorporating ophthalmic assessments, the application of AI to this field is still in its early stages. In this review article, we concentrate on the areas where AI‐based investigations could expand on existing conventional analyses to produce fresh findings using retinal biomarkers of systemic diseases. Five databases—Medline, Scopus, PubMed, Google Scholar, and Web of Science were searched using terms related to ocular imaging, systemic diseases, and artificial intelligence characteristics. Our review found that AI has been employed in a wide range of clinical tests and research applications, primarily for disease prediction, finding biomarkers and risk factor identification. We envisage artificial intelligence‐based models to have significant clinical and research impacts in the future through screening for high‐risk individuals, particularly in less developed areas, and identifying new retinal biomarkers, even though technical and socioeconomic challenges remain. Further research is needed to validate these models in real‐world setting.
非侵入性、快速和广泛使用的定量高分辨率成像方法的兴起，如现代视网膜摄影和光学相干断层扫描(OCT)，对眼科产生了重大影响。这些技术在评估眼部疾病方面提供了显著的准确性和分辨率，并越来越多地认识到它们在识别全身性疾病的眼部生物标志物方面的潜力。人工智能(AI)的应用已被证明在识别年龄、性别、收缩压、吸烟状况以及从眼底和OCT图像评估心血管疾病方面具有良好的效果。虽然我们对眼身关系的理解已经从几十年来基于大量人群的传统统计建模研究中得到了发展，但人工智能在这一领域的应用仍处于早期阶段。在这篇综述文章中，我们集中讨论了基于人工智能的研究可以扩展现有传统分析的领域，从而利用系统性疾病的视网膜生物标志物产生新的发现。五个数据库- medline, Scopus, PubMed，谷歌Scholar和Web of Science使用与眼部成像，全身性疾病和人工智能特征相关的术语进行了搜索。我们的审查发现，人工智能已广泛应用于临床试验和研究应用，主要用于疾病预测、寻找生物标志物和风险因素识别。尽管技术和社会经济挑战依然存在，但我们设想基于人工智能的模型通过筛查高风险个体，特别是在欠发达地区，以及识别新的视网膜生物标志物，在未来对临床和研究产生重大影响。需要进一步的研究来验证这些模型在现实世界的设置。
{"title":"Use of artificial intelligence algorithms to predict systemic diseases from retinal images","authors":"R. Khan, Janani Surya, Maitreyee Roy, M. S. Swathi Priya, Sashwanthi Mohan, S. Raman, Akshay Raman, Abhishek Vyas, R. Raman","doi":"10.1002/widm.1506","DOIUrl":"https://doi.org/10.1002/widm.1506","url":null,"abstract":"The rise of non‐invasive, rapid, and widely accessible quantitative high‐resolution imaging methods, such as modern retinal photography and optical coherence tomography (OCT), has significantly impacted ophthalmology. These techniques offer remarkable accuracy and resolution in assessing ocular diseases and are increasingly recognized for their potential in identifying ocular biomarkers of systemic diseases. The application of artificial intelligence (AI) has been demonstrated to have promising results in identifying age, gender, systolic blood pressure, smoking status, and assessing cardiovascular disorders from the fundus and OCT images. Although our understanding of eye–body relationships has advanced from decades of conventional statistical modeling in large population‐based studies incorporating ophthalmic assessments, the application of AI to this field is still in its early stages. In this review article, we concentrate on the areas where AI‐based investigations could expand on existing conventional analyses to produce fresh findings using retinal biomarkers of systemic diseases. Five databases—Medline, Scopus, PubMed, Google Scholar, and Web of Science were searched using terms related to ocular imaging, systemic diseases, and artificial intelligence characteristics. Our review found that AI has been employed in a wide range of clinical tests and research applications, primarily for disease prediction, finding biomarkers and risk factor identification. We envisage artificial intelligence‐based models to have significant clinical and research impacts in the future through screening for high‐risk individuals, particularly in less developed areas, and identifying new retinal biomarkers, even though technical and socioeconomic challenges remain. Further research is needed to validate these models in real‐world setting.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88036615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rule‐based systems have been used in the legal domain since the 1970s. Save for rare exceptions, machine learning has only recently been used. But why this delay? We investigate the appropriate use of machine learning to support and make legal predictions. To do so, we need to examine the appropriate use of data in global legal domains—including in common law, civil law, and hybrid jurisdictions. The use of various forms of Artificial Intelligence, including rule‐based reasoning, case‐based reasoning and machine learning in law requires an understanding of jurisprudential theories. We will see that the use of machine learning is particularly appropriate for non‐professionals: in particular self‐represented litigants or those relying upon legal aid services. The primary use of machine learning to support decision‐making in legal domains has been in criminal detection, financial domains, and sentencing. The use in these areas has led to concerns that the inappropriate use of Artificial Intelligence leads to biased decision making. This requires us to examine concerns about governance and ethics. Ethical concerns can be minimized by providing enhanced explanation, choosing appropriate data to be used, appropriately cleaning that data, and having human reviews of any decisions.
{"title":"The benefits and dangers of using machine learning to support making legal predictions","authors":"John Zeleznikow","doi":"10.1002/widm.1505","DOIUrl":"https://doi.org/10.1002/widm.1505","url":null,"abstract":"Rule‐based systems have been used in the legal domain since the 1970s. Save for rare exceptions, machine learning has only recently been used. But why this delay? We investigate the appropriate use of machine learning to support and make legal predictions. To do so, we need to examine the appropriate use of data in global legal domains—including in common law, civil law, and hybrid jurisdictions. The use of various forms of Artificial Intelligence, including rule‐based reasoning, case‐based reasoning and machine learning in law requires an understanding of jurisprudential theories. We will see that the use of machine learning is particularly appropriate for non‐professionals: in particular self‐represented litigants or those relying upon legal aid services. The primary use of machine learning to support decision‐making in legal domains has been in criminal detection, financial domains, and sentencing. The use in these areas has led to concerns that the inappropriate use of Artificial Intelligence leads to biased decision making. This requires us to examine concerns about governance and ethics. Ethical concerns can be minimized by providing enhanced explanation, choosing appropriate data to be used, appropriately cleaning that data, and having human reviews of any decisions.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80828066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-01Epub Date: 2023-02-19DOI: 10.1002/widm.1492
Elizabeth Ford, Richard Milne, Keegan Curlewis
Dementia poses a growing challenge for health services but remains stigmatized and under-recognized. Digital technologies to aid the earlier detection of dementia are approaching market. These include traditional cognitive screening tools presented on mobile devices, smartphone native applications, passive data collection from wearable, in-home and in-car sensors, as well as machine learning techniques applied to clinic and imaging data. It has been suggested that earlier detection and diagnosis may help patients plan for their future, achieve a better quality of life, and access clinical trials and possible future disease modifying treatments. In this review, we explore whether digital tools for the early detection of dementia can or should be deployed, by assessing them against the principles of ethical screening programs. We conclude that while the importance of dementia as a health problem is unquestionable, significant challenges remain. There is no available treatment which improves the prognosis of diagnosed disease. Progression from early-stage disease to dementia is neither given nor currently predictable. Available technologies are generally not both minimally invasive and highly accurate. Digital deployment risks exacerbating health inequalities due to biased training data and inequity in digital access. Finally, the acceptability of early dementia detection is not established, and resources would be needed to ensure follow-up and support for those flagged by any new system. We conclude that early dementia detection deployed at scale via digital technologies does not meet standards for a screening program and we offer recommendations for moving toward an ethical mode of implementation. This article is categorized under:Application Areas > Health CareCommercial, Legal, and Ethical Issues > Ethical ConsiderationsTechnologies > Artificial Intelligence.
{"title":"Ethical issues when using digital biomarkers and artificial intelligence for the early detection of dementia.","authors":"Elizabeth Ford, Richard Milne, Keegan Curlewis","doi":"10.1002/widm.1492","DOIUrl":"10.1002/widm.1492","url":null,"abstract":"<p><p>Dementia poses a growing challenge for health services but remains stigmatized and under-recognized. Digital technologies to aid the earlier detection of dementia are approaching market. These include traditional cognitive screening tools presented on mobile devices, smartphone native applications, passive data collection from wearable, in-home and in-car sensors, as well as machine learning techniques applied to clinic and imaging data. It has been suggested that earlier detection and diagnosis may help patients plan for their future, achieve a better quality of life, and access clinical trials and possible future disease modifying treatments. In this review, we explore whether digital tools for the early detection of dementia can or should be deployed, by assessing them against the principles of ethical screening programs. We conclude that while the importance of dementia as a health problem is unquestionable, significant challenges remain. There is no available treatment which improves the prognosis of diagnosed disease. Progression from early-stage disease to dementia is neither given nor currently predictable. Available technologies are generally not both minimally invasive and highly accurate. Digital deployment risks exacerbating health inequalities due to biased training data and inequity in digital access. Finally, the acceptability of early dementia detection is not established, and resources would be needed to ensure follow-up and support for those flagged by any new system. We conclude that early dementia detection deployed at scale via digital technologies does not meet standards for a screening program and we offer recommendations for moving toward an ethical mode of implementation. This article is categorized under:Application Areas > Health CareCommercial, Legal, and Ethical Issues > Ethical ConsiderationsTechnologies > Artificial Intelligence.</p>","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10909482/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75537809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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}
Indrajeet Ghosh, Sreenivasan Ramasamy Ramamurthy, Avijoy Chakma, Nirmalya Roy
The rapid and impromptu interest in the coupling of machine learning (ML) algorithms with wearable and contactless sensors aimed at tackling real‐world problems warrants a pedagogical study to understand all the aspects of this research direction. Considering this aspect, this survey aims to review the state‐of‐the‐art literature on ML algorithms, methodologies, and hypotheses adopted to solve the research problems and challenges in the domain of sports. First, we categorize this study into three main research fields: sensors, computer vision, and wireless and mobile‐based applications. Then, for each of these fields, we thoroughly analyze the systems that are deployable for real‐time sports analytics. Next, we meticulously discuss the learning algorithms (e.g., statistical learning, deep learning, reinforcement learning) that power those deployable systems while also comparing and contrasting the benefits of those learning methodologies. Finally, we highlight the possible future open‐research opportunities and emerging technologies that could contribute to the domain of sports analytics.
{"title":"Sports analytics review: Artificial intelligence applications, emerging technologies, and algorithmic perspective","authors":"Indrajeet Ghosh, Sreenivasan Ramasamy Ramamurthy, Avijoy Chakma, Nirmalya Roy","doi":"10.1002/widm.1496","DOIUrl":"https://doi.org/10.1002/widm.1496","url":null,"abstract":"The rapid and impromptu interest in the coupling of machine learning (ML) algorithms with wearable and contactless sensors aimed at tackling real‐world problems warrants a pedagogical study to understand all the aspects of this research direction. Considering this aspect, this survey aims to review the state‐of‐the‐art literature on ML algorithms, methodologies, and hypotheses adopted to solve the research problems and challenges in the domain of sports. First, we categorize this study into three main research fields: sensors, computer vision, and wireless and mobile‐based applications. Then, for each of these fields, we thoroughly analyze the systems that are deployable for real‐time sports analytics. Next, we meticulously discuss the learning algorithms (e.g., statistical learning, deep learning, reinforcement learning) that power those deployable systems while also comparing and contrasting the benefits of those learning methodologies. Finally, we highlight the possible future open‐research opportunities and emerging technologies that could contribute to the domain of sports analytics.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79200830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alex Gaudio, C. Faloutsos, A. Smailagic, P. Costa, A. Campilho
Is there an initialization for deep networks that requires no learning? ExplainFix adopts two design principles: the “fixed filters” principle that all spatial filter weights of convolutional neural networks can be fixed at initialization and never learned, and the “nimbleness” principle that only few network parameters suffice. We contribute (a) visual model‐based explanations, (b) speed and accuracy gains, and (c) novel tools for deep convolutional neural networks. ExplainFix gives key insights that spatially fixed networks should have a steered initialization, that spatial convolution layers tend to prioritize low frequencies, and that most network parameters are not necessary in spatially fixed models. ExplainFix models have up to ×100 fewer spatial filter kernels than fully learned models and matching or improved accuracy. Our extensive empirical analysis confirms that ExplainFix guarantees nimbler models (train up to 17% faster with channel pruning), matching or improved predictive performance (spanning 13 distinct baseline models, four architectures and two medical image datasets), improved robustness to larger learning rate, and robustness to varying model size. We are first to demonstrate that all spatial filters in state‐of‐the‐art convolutional deep networks can be fixed at initialization, not learned.
{"title":"ExplainFix: Explainable spatially fixed deep networks","authors":"Alex Gaudio, C. Faloutsos, A. Smailagic, P. Costa, A. Campilho","doi":"10.1002/widm.1483","DOIUrl":"https://doi.org/10.1002/widm.1483","url":null,"abstract":"Is there an initialization for deep networks that requires no learning? ExplainFix adopts two design principles: the “fixed filters” principle that all spatial filter weights of convolutional neural networks can be fixed at initialization and never learned, and the “nimbleness” principle that only few network parameters suffice. We contribute (a) visual model‐based explanations, (b) speed and accuracy gains, and (c) novel tools for deep convolutional neural networks. ExplainFix gives key insights that spatially fixed networks should have a steered initialization, that spatial convolution layers tend to prioritize low frequencies, and that most network parameters are not necessary in spatially fixed models. ExplainFix models have up to ×100 fewer spatial filter kernels than fully learned models and matching or improved accuracy. Our extensive empirical analysis confirms that ExplainFix guarantees nimbler models (train up to 17% faster with channel pruning), matching or improved predictive performance (spanning 13 distinct baseline models, four architectures and two medical image datasets), improved robustness to larger learning rate, and robustness to varying model size. We are first to demonstrate that all spatial filters in state‐of‐the‐art convolutional deep networks can be fixed at initialization, not learned.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73146059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alex Gaudio, A. Smailagic, C. Faloutsos, Shreshta Mohan, Elvin Johnson, Yuhao Liu, P. Costa, A. Campilho
Explanations of a model's biases or predictions are essential to medical image analysis. Yet, explainable machine learning approaches for medical image analysis are challenged by needs to preserve privacy of patient data, and by current trends in deep learning to use unsustainably large models and large datasets. We propose DeepFixCX for explainable and privacy‐preserving medical image compression that is nimble and performant. We contribute a review of the field and a conceptual framework for simultaneous privacy and explainability via tools of compression. DeepFixCX compresses images without learning by removing or obscuring spatial and edge information. DeepFixCX is ante‐hoc explainable and gives privatized post hoc explanations of spatial and edge bias without accessing the original image. DeepFixCX privatizes images to prevent image reconstruction and mitigate patient re‐identification. DeepFixCX is nimble. Compression can occur on a laptop CPU or GPU to compress and privatize 1700 images per second of size 320 × 320. DeepFixCX enables use of low memory MLP classifiers for vision data; permitting small performance loss gives end‐to‐end MLP performance over 70× faster and batch size over 100× larger. DeepFixCX consistently improves predictive classification performance of a Deep Neural Network (DNN) by 0.02 AUC ROC on Glaucoma and Cervix Type detection datasets, and can improve multi‐label chest x‐ray classification performance in seven of 10 tested settings. In all three datasets, compression to less than 5% of original number of pixels gives matching or improved performance. Our main novelty is to define an explainability versus privacy problem and address it with lossy compression.
{"title":"DeepFixCX: Explainable privacy‐preserving image compression for medical image analysis","authors":"Alex Gaudio, A. Smailagic, C. Faloutsos, Shreshta Mohan, Elvin Johnson, Yuhao Liu, P. Costa, A. Campilho","doi":"10.1002/widm.1495","DOIUrl":"https://doi.org/10.1002/widm.1495","url":null,"abstract":"Explanations of a model's biases or predictions are essential to medical image analysis. Yet, explainable machine learning approaches for medical image analysis are challenged by needs to preserve privacy of patient data, and by current trends in deep learning to use unsustainably large models and large datasets. We propose DeepFixCX for explainable and privacy‐preserving medical image compression that is nimble and performant. We contribute a review of the field and a conceptual framework for simultaneous privacy and explainability via tools of compression. DeepFixCX compresses images without learning by removing or obscuring spatial and edge information. DeepFixCX is ante‐hoc explainable and gives privatized post hoc explanations of spatial and edge bias without accessing the original image. DeepFixCX privatizes images to prevent image reconstruction and mitigate patient re‐identification. DeepFixCX is nimble. Compression can occur on a laptop CPU or GPU to compress and privatize 1700 images per second of size 320 × 320. DeepFixCX enables use of low memory MLP classifiers for vision data; permitting small performance loss gives end‐to‐end MLP performance over 70× faster and batch size over 100× larger. DeepFixCX consistently improves predictive classification performance of a Deep Neural Network (DNN) by 0.02 AUC ROC on Glaucoma and Cervix Type detection datasets, and can improve multi‐label chest x‐ray classification performance in seven of 10 tested settings. In all three datasets, compression to less than 5% of original number of pixels gives matching or improved performance. Our main novelty is to define an explainability versus privacy problem and address it with lossy compression.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76416264","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Interpretability and explainability are crucial for machine learning (ML) and statistical applications in medicine, economics, law, and natural sciences and form an essential principle for ML model design and development. Although interpretability and explainability have escaped a precise and universal definition, many models and techniques motivated by these properties have been developed over the last 30 years, with the focus currently shifting toward deep learning. We will consider concrete examples of state‐of‐the‐art, including specially tailored rule‐based, sparse, and additive classification models, interpretable representation learning, and methods for explaining black‐box models post hoc. The discussion will emphasize the need for and relevance of interpretability and explainability, the divide between them, and the inductive biases behind the presented “zoo” of interpretable models and explanation methods.
{"title":"Interpretable and explainable machine learning: A methods‐centric overview with concrete examples","authors":"Ricards Marcinkevics, Julia E. Vogt","doi":"10.1002/widm.1493","DOIUrl":"https://doi.org/10.1002/widm.1493","url":null,"abstract":"Interpretability and explainability are crucial for machine learning (ML) and statistical applications in medicine, economics, law, and natural sciences and form an essential principle for ML model design and development. Although interpretability and explainability have escaped a precise and universal definition, many models and techniques motivated by these properties have been developed over the last 30 years, with the focus currently shifting toward deep learning. We will consider concrete examples of state‐of‐the‐art, including specially tailored rule‐based, sparse, and additive classification models, interpretable representation learning, and methods for explaining black‐box models post hoc. The discussion will emphasize the need for and relevance of interpretability and explainability, the divide between them, and the inductive biases behind the presented “zoo” of interpretable models and explanation methods.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75749279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 ever‐growing adoption of artificial intelligence (AI)‐based systems, the carbon footprint of AI is no longer negligible. AI researchers and practitioners are therefore urged to hold themselves accountable for the carbon emissions of the AI models they design and use. This led in recent years to the appearance of researches tackling AI environmental sustainability, a field referred to as Green AI. Despite the rapid growth of interest in the topic, a comprehensive overview of Green AI research is to date still missing. To address this gap, in this article, we present a systematic review of the Green AI literature. From the analysis of 98 primary studies, different patterns emerge. The topic experienced a considerable growth from 2020 onward. Most studies consider monitoring AI model footprint, tuning hyperparameters to improve model sustainability, or benchmarking models. A mix of position papers, observational studies, and solution papers are present. Most papers focus on the training phase, are algorithm‐agnostic or study neural networks, and use image data. Laboratory experiments are the most common research strategy. Reported Green AI energy savings go up to 115%, with savings over 50% being rather common. Industrial parties are involved in Green AI studies, albeit most target academic readers. Green AI tool provisioning is scarce. As a conclusion, the Green AI research field results to have reached a considerable level of maturity. Therefore, from this review emerges that the time is suitable to adopt other Green AI research strategies, and port the numerous promising academic results to industrial practice.
{"title":"A systematic review of Green AI","authors":"R. Verdecchia, June Sallou, Luís Cruz","doi":"10.1002/widm.1507","DOIUrl":"https://doi.org/10.1002/widm.1507","url":null,"abstract":"With the ever‐growing adoption of artificial intelligence (AI)‐based systems, the carbon footprint of AI is no longer negligible. AI researchers and practitioners are therefore urged to hold themselves accountable for the carbon emissions of the AI models they design and use. This led in recent years to the appearance of researches tackling AI environmental sustainability, a field referred to as Green AI. Despite the rapid growth of interest in the topic, a comprehensive overview of Green AI research is to date still missing. To address this gap, in this article, we present a systematic review of the Green AI literature. From the analysis of 98 primary studies, different patterns emerge. The topic experienced a considerable growth from 2020 onward. Most studies consider monitoring AI model footprint, tuning hyperparameters to improve model sustainability, or benchmarking models. A mix of position papers, observational studies, and solution papers are present. Most papers focus on the training phase, are algorithm‐agnostic or study neural networks, and use image data. Laboratory experiments are the most common research strategy. Reported Green AI energy savings go up to 115%, with savings over 50% being rather common. Industrial parties are involved in Green AI studies, albeit most target academic readers. Green AI tool provisioning is scarce. As a conclusion, the Green AI research field results to have reached a considerable level of maturity. Therefore, from this review emerges that the time is suitable to adopt other Green AI research strategies, and port the numerous promising academic results to industrial practice.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82554190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Data mining (DM) and machine learning (ML) applications in medical diagnostic systems are budding. Data privacy is essential in these systems as healthcare data are highly sensitive. The proposed work first discusses various privacy and security challenges in these systems. To address these next, we discuss different privacy‐preserving (PP) computation techniques in the context of DM and ML for secure data evaluation and processing. The state‐of‐the‐art applications of these systems in healthcare are analyzed at various stages such as data collection, data publication, data distribution, and output phases regarding PPDM and input, model, training, and output phases in the context of PPML. Furthermore, PP federated learning is also discussed. Finally, we present open challenges in these systems and future research directions.
{"title":"Privacy‐preserving data mining and machine learning in healthcare: Applications, challenges, and solutions","authors":"V. Naresh, Muthusamy Thamarai","doi":"10.1002/widm.1490","DOIUrl":"https://doi.org/10.1002/widm.1490","url":null,"abstract":"Data mining (DM) and machine learning (ML) applications in medical diagnostic systems are budding. Data privacy is essential in these systems as healthcare data are highly sensitive. The proposed work first discusses various privacy and security challenges in these systems. To address these next, we discuss different privacy‐preserving (PP) computation techniques in the context of DM and ML for secure data evaluation and processing. The state‐of‐the‐art applications of these systems in healthcare are analyzed at various stages such as data collection, data publication, data distribution, and output phases regarding PPDM and input, model, training, and output phases in the context of PPML. Furthermore, PP federated learning is also discussed. Finally, we present open challenges in these systems and future research directions.","PeriodicalId":48970,"journal":{"name":"Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery","volume":null,"pages":null},"PeriodicalIF":7.8,"publicationDate":"2023-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84036376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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：