A kind of text information retrieval method and related device

Abstract

The present application provides a text information retrieval method and a related device, which can improve the retrieval success rate. In this method, after the text information to be retrieved is obtained, the keyword set is first obtained according to the target word set included in the text information to be retrieved. Then an aggregation keyword set is generated according to the keyword set, and the aggregation keyword set includes a target aggregation keyword sub-set and at least one aggregation keyword sub-set. Then generate the first target feature vector corresponding to the target aggregation keyword sub-set according to the aggregation keyword set, wherein the first target feature vector is generated based on the original feature vector and the discrimination feature vector, the original feature vector and the target aggregation keyword The sub-sets have an association relationship, and the discriminative feature vector has an association relationship with at least one aggregation key sub-set. Finally, the retrieval score corresponding to the first object feature vector is obtained through the object classification model, wherein the retrieval score is used to determine the retrieval result.

Description

technical field [0001] The present application relates to the field of network technology, in particular to a text information retrieval method and a related device. Background technique [0002] With the rapid development of network technology, users can search information on the network. Specifically, when a user uses software with a retrieval function on a terminal device, he can input text information to be retrieved, and the software can intelligently retrieve the text information and return corresponding retrieval results. [0003] In the above-mentioned intelligent retrieval process, it is usually necessary to determine the keywords in the text information first. For example, when the text information is where Zhang XX was born, the subject of the text information can be determined first as a person, and the predicate associated with the person can be obtained, such as birth location, family members, and work place, etc., and then calculate the similarity between the...

AI Technical Summary

Problems solved by technology

Because a high similarity threshold is usually set, some text information cannot be retrieved, resulting in a low retrieval success rate

Images