Short text recommendation method and device, medium and electronic equipment

Abstract

The invention provides a short text recommendation method and device, a medium and electronic equipment. The short text recommendation method comprises the following steps: acquiring a first short text input by a user; obtaining a plurality of second short texts in an ES mode, obtaining a semantic vector of the first short text and a semantic vector of each second short text, calculating the similarity between the semantic vector of each second short text and the semantic vector of the first short text, wherein the semantic vector of the short text is used for representing the feature vector of the short text semantics; and finally sorting the plurality of second short texts according to the similarity between the semantic vector of each second short text and the semantic vector of the first short text, and recommending the sorted second short texts. According to the scheme, the short texts are recalled and sorted from the deep semantic level of the short texts, so that the actual requirements of users can be better met, and the click rate of recommended contents is effectively increased.

Description

technical field [0001] The invention relates to the technical field of big data analysis, in particular to a short text recommendation method, device, medium and electronic equipment. Background technique [0002] With the advent of the big data era, how to help users quickly obtain useful information from a large amount of information has become an important task for many businesses, and personalized recommendation systems have emerged as the times require. Based on massive data mining, personalized recommendation system guides users to find their own information needs. It has been widely used in many fields, especially short text search recommendation is the most widely used. [0003] At present, for online real-time short text text recommendation, there are mainly two types of technologies in the existing technology, one is to use distributed full-text search (ElasticSearch, ES) to retrieve and recall a specified number of short texts as recommended texts for users; the o...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a method, device, medium, and electronic device for recommending short texts, so as to solve the problem of directly displaying recalled short texts or sorting and recommending according to keywords in the prior art. The short texts that users can view earlier may not be The content that is closest to the actual needs of users leads to the problem of low click-through rate

Images