Knowledge Graphs Aided Entity Relation Networks Using Chinese Quality Supervision News
Keywords:
Open information extraction, Knowledge bases, Relation extraction, Graph neural networksAbstract
This paper, with the aid of knowledge bases and graph neural networks, tries a new way to information extraction and correlation analysis in the Chinese open domain. Chinese quality supervision news are taken as the corpus. We categorize event types based on topics generated via LDA. As for entities in texts, their types and relations are recognized through a combination of local and distant knowledge graphs. Those relations that do not exist in the knowledge graphs can be predicted through graph neural networks, with fully connected dependency syntactic trees as inputs. The correlation analysis on entities and events from quality news provides supports for relevant departments of the government, manufacturers, and consumers.
References
TSENG Y H, LEE L H, LIN S Y and et al. Chinese open relation extraction for knowledge acquisition. In Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics,2014. pp:12-16.
MINTZ M, BILLS S, SNOW R, and et al. Distant supervision for relation extraction without labeled data. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP. 2009:1003- 1011.
JI G L, LIU K, HE S Z, et al. Distant supervision for relation extraction with sentence-level attention and entity descriptions. In Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. 2017:3060-3066.
LIN Y K, SHEN S Q, LIU Z Y. Neural relation extraction with selective attention over instances. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2016:2124-2133.
