Entity linking method and device, equipment and storage medium

Abstract

The invention discloses an entity linking method and device, equipment and a storage medium, and relates to the field of natural language process. The method comprises: obtaining to-be-linked entities and context content corresponding to the to-be-linked entities from a to-be-identified text, and, according to the to-be-linked entities, determining a plurality of candidate entities and entity information of each candidate entity in the plurality of candidate entities in the knowledge graph; matching the context content with the entity information corresponding to each candidate entity to obtain a first matching score, and determining a target candidate entity in the plurality of candidate entities according to the first matching score; and based on the disambiguation score between the to-be-linked entity and the target candidate entity, determining a target entity in the target candidate entity, and determining the target entity as a linked entity corresponding to the to-be-linked entity.

Description

technical field [0001] The present application relates to the field of natural language processing, and in particular to a method, device, device and storage medium for entity linking. Background technique [0002] Entity linking (EL) technology is a hot spot in the field of natural language processing in recent years, especially in scenarios such as knowledge graph construction, which plays a very important role. Specifically, entity linking is a technology that maps the entities appearing in the text to be recognized to a given knowledge graph, and is used to match the entities to be recognized with the entities existing in the knowledge graph to complete question answering, semantic search, etc. , information extraction and other natural language tasks. [0003] Entity linking can include two processes of entity recognition and entity disambiguation. Entity recognition is used to determine multiple candidate entities in the knowledge graph according to the entity to be r...

AI Technical Summary

Problems solved by technology

[0004] Since the entity disambiguation method based on the pre-trained language model will consider the entity information of all candidate entities in the knowledge graph, this will cause entity disambiguation to cover up more relevant information because too much entity information is considered, and when there are too many candidate entities, This entity disambiguation method requires a huge amount of calculation and takes a long time due to the need to perform disambiguation scores on all candidate entities

Images