Named entity recognition method for conditional random field based on word vector representation

Abstract

The invention provides a named entity recognition method for a conditional random field based on word vector representation and belongs to the field of natural language processing technology. A conditional random field algorithm based on word vector representation, a conditional random field algorithm adopting fusion word vector representation and an online named entity recognition system adopting B / S physical design and providing a graphic interaction interface are included in the method. Through the method, a biomedical named entity of a biomedical text to be analyzed by a user can be recognized, semantic representation characteristics of word vectors are brought into play in the recognition process with little dependency on artificial feature participation, the problem that the conditional random field is only valid to discrete feature representation is solved, and the advantages of the conditional random field algorithm, which is a discriminant undirected graph model, are also brought into play; the method provides named entity interactive relationship data retrieval service for the user and also provides a correction function on an automatic analysis result for the user.

Description

technical field [0001] The invention belongs to the field of natural language processing, and relates to a method for high-quality biological named entity recognition on biomedical texts, in particular to a biological named entity recognition method based on the fusion of a conditional random field (CRF) model and a word representation method. Background technique [0002] The task of named entity recognition is to identify words or phrases with specific meanings such as names of people, places, and institutions that appear in the text. Named entity recognition in the field of biomedicine is called Biomedical Named Entity Recognition (Bio-NER), which aims to use biomedical text mining technology to identify specific types of entity names that appear in biomedical literature, such as protein , genes, diseases, cells, etc. are automatically identified and classified. Biological named entity recognition is a key step in biomedical text mining and a prerequisite for deep text m...

AI Technical Summary

Problems solved by technology

[0005] The invention provides a named entity recognition method based on a conditional random field represented by a word vector, which firstly solves the problems of high cost and low generalization ability caused by manual feature extraction, and secondly solves the problem that the conditional random field is only suitable for discrete The problem of effective feature representation, finally, the accuracy of conditional random field for named entity recognition is improved due to the integration of semantic representation of word vectors

Images