Person-relation extraction using bert based knowledge graph

Abstract

Artificial intelligence technology has been actively researched in the areas of image processing and natural language processing. Recently, with the release of Google’s language model BERT, the importance of artificial intelligence models has attracted attention in the field of natural language processing. In this paper, we propose a knowledge graph to build a model that can extract people in a document using BERT, and to grasp the relationship between people based on the model. In addition, to verify the applicability of person extraction techniques using BERT based knowledge graphs, we conduct a performance comparison experiment with other person extraction models and apply our proposed method to the case study. © 2020, ICIC International.