[Objective] This paper reviews the related researches and resources in the field of named entity disambiguation (NED) focusing on the research progress of NED methods.
[Coverage] We reviewed a total of 57 papers and electronic resources from CNKI, Wanfang Data Knowledge Service Platform, and EBSCO.
[Methods] We first teases out the available thought and method for NED in terms of entity prominence, context similarity, entity relationship, deep learning and special identification resources. Next, some useful knowledge bases, open source tools as well as international conferences on NED evaluation are listed and discussed here.
[Results] Traditional methods are classic and easy to use, while new methods such as deep learning, which have emerged in recent years, have significantly improved the disambiguation effect. Effective disambiguation models often integrate different types of methods in order to achieve the optimal disambiguation effect.
[Limitations] There is still subjectivity in the comparative analysis of different methods based on the existing literature.
[Conclusions] The existing NED methods are still in the development stage. Artificial intelligence methods and field resources can be used to further improve the entity disambiguation effect in the future.
温萍梅, 叶志炜, 丁文健, 刘颖, 徐健. 命名实体消歧研究进展综述
[J]. 数据分析与知识发现, 0, (): 1-.
Wen Pingmei, Ye Zhiwei, Ding Wenjian, Liu Ying, Xu Jian. Literature Review on the Progress of Named Entity Disambiguation Research
. Data Analysis and Knowledge Discovery, 0, (): 1-.