Named entity identification method based on knowledge graph

Abstract

The invention discloses a named entity recognition method based on a knowledge graph; the method carries out the reasoning of a relation indicator r and a tail entity t based on entity relation triples (h, r, t) in an existing knowledge graph. On the basis, a new model is provided to be used for representation learning among triple entity relationships extracted from an open domain knowledge base,so that relationship mining and knowledge reasoning in a knowledge graph are realized, and the effects of optimizing association search and personalized recommendation effects are achieved. Comparedwith a traditional knowledge base, the open domain knowledge base uses the relation indicator words to replace the relation types, entities are richer, and the granularity is finer and smoother.

Description

technical field [0001] The present invention relates to a named entity recognition method, in particular to a named entity recognition method based on a knowledge map. Background technique [0002] In the past ten years, with the continuous development of artificial intelligence, cloud computing and other technologies, the construction of large-scale knowledge bases has made good progress, and has been widely used in e-commerce, cultural entertainment, finance, logistics and other industries. In a number of businesses, good results have been achieved. [0003] "Named Entity", which is widely used in the field of natural language processing, was first proposed in 1996 at the Sixth Information Understanding Conference (MUC-6), mainly to extract specific information such as company activities and national defense activities from a given text , the text can be structured, semi-structured, or unstructured data. When performing information extraction tasks, some entities with sp...

AI Technical Summary

Problems solved by technology

"JointExtraction of Entities and Relations Based on a Novel Tagging Scheme" published by Suncong Zheng et al. in 2017 proposed a new method for entity recognition and relationship extraction. The traditional NER and RC tasks are separated, thus ignoring the The association between entities and relationships, and errors in entity extraction annotations will implicate relationship extraction

However, these algorithms are not well suited for knowledge reasoning. As the data scale of the knowledge base continues to expand, these algorithms based on mesh representations have the following two problems: 1. Insufficient computational efficiency; 2. Cannot obtain sparse data good feedback

Knowledge reasoning based on network structure knowledge base is difficult to better meet the needs of real-time computing

Symbol-based knowledge bases in the form of networks cannot cope with numerical calculations in continuous spaces

Simple symbols and logical representations make the knowledge in the knowledge base more and more discrete, and the knowledge cannot be well integrated, which also makes it impossible for the intelligent system to use the knowledge base more flexibly for knowledge reasoning

Images