Explainable artificial intelligence: an analytical review

IF 6.4 2区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery Pub Date : 2021-07-12 DOI:10.1002/widm.1424

P. Angelov, E. Soares, Richard Jiang, Nicholas I. Arnold, Peter M. Atkinson

引用次数: 208

Abstract

This paper provides a brief analytical review of the current state‐of‐the‐art in relation to the explainability of artificial intelligence in the context of recent advances in machine learning and deep learning. The paper starts with a brief historical introduction and a taxonomy, and formulates the main challenges in terms of explainability building on the recently formulated National Institute of Standards four principles of explainability. Recently published methods related to the topic are then critically reviewed and analyzed. Finally, future directions for research are suggested.

查看原文本刊更多论文

可解释的人工智能:分析回顾

本文在机器学习和深度学习的最新进展背景下，对人工智能的可解释性进行了简要的分析回顾。本文从简要的历史介绍和分类开始，并根据最近制定的国家标准研究所可解释性的四项原则，阐述了可解释性方面的主要挑战。最近发表的方法相关的主题，然后严格审查和分析。最后，对今后的研究方向提出了建议。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE-COMPUTER SCIENCE, THEORY & METHODS

CiteScore

22.70

自引率

2.60%

发文量

审稿时长

>12 weeks

期刊介绍： The goals of Wiley Interdisciplinary Reviews-Data Mining and Knowledge Discovery (WIREs DMKD） are multifaceted. Firstly, the journal aims to provide a comprehensive overview of the current state of data mining and knowledge discovery by featuring ongoing reviews authored by leading researchers. Secondly, it seeks to highlight the interdisciplinary nature of the field by presenting articles from diverse perspectives, covering various application areas such as technology, business, healthcare, education, government, society, and culture. Thirdly, WIREs DMKD endeavors to keep pace with the rapid advancements in data mining and knowledge discovery through regular content updates. Lastly, the journal strives to promote active engagement in the field by presenting its accomplishments and challenges in an accessible manner to a broad audience. The content of WIREs DMKD is intended to benefit upper-level undergraduate and postgraduate students, teaching and research professors in academic programs, as well as scientists and research managers in industry.