Information search method, device, storage medium and electronic device

Abstract

The present disclosure relates to an information search method, device, storage medium, and electronic equipment. The method includes: determining a sequence of phrases contained in a search character string; taking each phrase in the sequence of phrases as a target phrase, and targeting each of the phrases The target phrase performs the following operations: using the target phrase as a keyword, determining a search entity corresponding to the keyword; determining the historical relevance between the target phrase and the search entity according to historical search data; Context information relevancy between other phrases except the target phrase and the search entity; sort each search entity according to the historical relevancy and the context information relevancy, and display according to the sorting result Informational search results for the search string. It is used to solve the technical problem of low accuracy rate of matched entities when entity matching is performed on search words by using entity linking technology in related technologies.

Description

technical field [0001] The present disclosure relates to the technical field of information processing, and in particular, relates to an information search method, device, storage medium and electronic equipment. Background technique [0002] In the related technology, in order to match the appropriate entity when searching for the target entity (entity) by keyword (query), an entity linking (entity linking) technology is adopted, which uses offline The mined entity-mention (mention-entity) data obtains the candidate entity set, and then combines the language model (language model) or semantic model (semantic model) to sort the candidate results to obtain the final entity link result. [0003] However, this entity link technology relies more on the NER (Named Entity Recognition) recognition model, and the recognition accuracy of the NER recognition model depends on the labeled training data, and the NER recognition model is mainly used to recognize names of people, places an...

AI Technical Summary

Problems solved by technology

[0003] However, this entity link technology relies more on the NER (Named Entity Recognition) recognition model, and the recognition accuracy of the NER recognition model depends on the labeled training data, and the NER recognition model is mainly used to recognize names of people, places and institutions. Emerging entity names have low recognition accuracy, which in turn leads to low accuracy matching entities for related search terms

Images