Recognition method and device for entities in dialogue corpus and computer equipment

Abstract

The embodiment of the invention discloses a recognition method and device for entities in dialogue corpora and computer equipment. The method comprises the steps that: the corpus text of entities to be recognized are obtained; the corpus text is subjected to word segmentation, a word segmentation result is obtained, wherein the word segmentation result comprises a plurality of characters; a character vector corresponding to each character in the word segmentation result is obtained, and the character vectors corresponding to the characters are combined, so that a text matrix corresponding to the corpus text is obtained; and the text matrix is taken as the input of an entity recognition model, and entities in the corpus text output by the entity recognition model are obtained. In this way,the accuracy of entity recognition can be improved.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular to a method, device, computer equipment and storage medium for identifying entities in dialogue materials. Background technique [0002] With the development of speech recognition technology, the bottleneck of speech recognition into text has been broken through, the meaning expressed by robots to people will be clearer, and the dialogue will be simpler. However, after speech recognition is performed on the speech, only a string of text is obtained, and the robot does not know the meaning expressed by the text. [0003] In order to understand the meaning of the text, the existing method is to identify the entities in the text, and then understand the meaning of the text according to the identified entities. However, the existing entity recognition models are usually trained based on the input word vectors to identify entities based on the input word information, wh...

AI Technical Summary

Problems solved by technology

However, the existing entity recognition models are usually trained based on the input word vectors to identify entities based on the input word information, which leads to low accuracy of the final recognized entities

Images