An Automatic Approach for Extracting Chinese Entity Synonyms from Encyclopedias
Subin Huang, Wei Qin, Senyan Zhao, Shengwei Gu · 2020
Synonyms play an important role in many entity-based applications. However, most known synonym extraction methods are in English, while Chinese ones are relatively rare. In this paper, we propose a simple yet effective extraction and cleaning framework that automatically extracts Chinese entity synonyms from encyclopedias. In the extraction phase, three extraction methods named direct extraction, pattern-based extraction and neural mining extraction are employed to acquire a large number of candidate entity synonym sets from multiple sources of Chinese online encyclopedias (e.g., title, infobox and abstract). In the cleaning phase, three error detection strategies named lexical and semantic rules, domain filtering and similarity filtering are used to improve the precision of the extracted candidate entity synonym sets. Hence, we build a Chinese synonym knowledge base named CN-EntitySyn. Experimental results show that CN-EntitySyn outperforms the compared baselines in terms of precision and coverage.