| 本文已被：浏览 6157次 下载 10748次 |
码上扫一扫！ |
|
|
| 知识图谱可解释推理研究综述 |
|
侯中妮1,2, 靳小龙1,2, 陈剑赟3, 官赛萍1, 王元卓1, 程学旗1,2
|
|
1.中国科学院网络数据科学与技术重点实验室(中国科学院 计算技术研究所), 北京 100190;2.中国科学院大学 计算机科学与技术学院, 北京 100049;3.北京市信息技术研究所, 北京 100094
|
|
| 摘要: |
| 面向知识图谱的知识推理旨在通过已有的知识图谱事实，去推断新的事实，进而实现知识库的补全.近年来，尽管基于分布式表示学习的方法在推理任务上取得了巨大的成功，但是他们的黑盒属性使得模型无法为预测出的事实做出解释.所以，如何设计用户可理解、可信赖的推理模型成为了人们关注的问题.从可解释性的基本概念出发，系统梳理了面向知识图谱的可解释知识推理的相关工作，具体介绍了事前可解释推理模型和事后可解释推理模型的研究进展；根据可解释范围的大小，将事前可解释推理模型进一步细分为全局可解释的推理和局部可解释的推理；在事后解释模型中，回顾了推理模型的代表方法，并详细介绍提供事后解释的两类解释方法.此外，还总结了可解释知识推理在医疗、金融领域的应用.随后，对可解释知识推理的现状进行概述，最后展望了可解释知识推理的未来发展方向，以期进一步推动可解释推理的发展和应用. |
| 关键词: 可解释性 知识推理 知识图谱 事后可解释 事前可解释 |
| DOI：10.13328/j.cnki.jos.006522 |
| 分类号: |
| 基金项目:国家自然科学基金（61772501，62002341，U1911401，U1836206）；国家重点研发计划（2018YFC0825205） |
|
| Survey of Interpretable Reasoning on Knowledge Graphs |
|
HOU Zhong-Ni1,2, JIN Xiao-Long1,2, CHEN Jian-Yun3, GUAN Sai-Ping1, WANG Yuan-Zhuo1, CHENG Xue-Qi1,2
|
|
1.CAS Key Laboratory of Network Data Science & Technology (Institute of Computing Technology, Chinese Academy of Sciences), Beijing 100190, China;2.School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 100049, China;3.Beijing Institute of Information Technology, Beijing 100094, China
|
| Abstract: |
| Reasoning over knowledge graphs aims to infer new facts based on known ones, so as to make the graphs as complete as possible. In recent years, distributed embedding-based reasoning methods have made great success on this task. However, due to their black-box nature, these methods cannot provide interpretability for a specific prediction. Therefore, there has been a growing interest in how to design user-understandable and user-trustworthy reasoning models. Starting from the basic concept of interpretability, this work systematically studies the recently developed methods for interpretable reasoning on knowledge graphs. Specifically, it introduces the research progress of ante-hoc and post-hoc interpretable reasoning models. According to the scope of interpretability, ante-hoc interpretable models can be further divided into local-interpretable and global-interpretable models. In post-hoc interpretable reasoning models, this study reviews representative reasoning methods and introduces two post-hoc interpretation methods in detail. Next, it also summarizes the application of explainable knowledge reasoning in such fields as finance and healthcare. Then, this study summarizes the current situation in explainable knowledge learning. Finally, the future technological development of interpretable reasoning models is prospected. |
| Key words: interpretability knowledge reasoning knowledge graph post-hoc interpretability ante-hoc interpretability |
