Data retrieval method and device

Abstract

The invention discloses a data retrieval method and device, and relates to the technical field of artificial intelligence, and the method comprises the steps: determining a plurality of query intentions according to a retrieval request of a user, wherein the query intention is used for representing a retrieval keyword corresponding to the semantic meaning of a user retrieval request; taking a plurality of query intentions corresponding to the retrieval request as input data, and inputting the input data into a query intention boundary recognition machine learning model to obtain a query intention boundary; classifying and sorting the plurality of query intentions according to the query intention boundary, and determining the plurality of query intentions in the query intention boundary range; according to the association relationship between the query intention of the user and the social network information of the user, screening and reordering the plurality of query intentions in the query intention boundary range; and determining a data retrieval result according to the plurality of screened and reordered query intentions in the query intention boundary range. According to the method, the data retrieval accuracy and retrieval efficiency can be improved, and the user experience is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a data retrieval method and device. Background technique [0002] This section is intended to provide a background or context to embodiments of the invention that are recited in the claims. The descriptions herein are not admitted to be prior art by inclusion in this section. [0003] At present, all mainstream APP applications have search functions, but most of them use keyword matching and word vector space to calculate similarity and display matching or similar results. Simply put, it is the semantics, logic and focus of text content. similarity. [0004] For example, the main method of related research on user retrieval is the classification of user search intentions and the determination of the relevance of results. There will also be more detailed classifications at the layer level, starting from the overall user's query goals, and studying the classifica...

AI Technical Summary

Problems solved by technology

[0006] However, the above-mentioned retrieval method is a unified retrieval of all query intentions, without considering the user's personalized retrieval needs, and cannot be adapted to different users, which may easily cause the user to find that the retrieval results do not meet expectations, and reduce the accuracy of information retrieval; At the same time, due to inaccurate retrieval results, users need to perform repeated retrievals, which also reduces the efficiency of information retrieval and easily causes bad user experience to users

Images