A search word recommendation method and device

Abstract

The embodiment of the present invention provides a method and device for recommending search words, which can obtain the browsing record of the target user; the browsing record contains at least one target browsing content identifier; based on the target browsing content identifier contained in the browsing record, and the pre-built browsing content Offline index of identification and search terms, recommending search terms for target users; offline index includes the corresponding relationship between each browsing content identification and a preset number of search terms, and the similarity between each browsing content identification and the corresponding search terms . Since the large amount of data of browsing records of many users is considered when building an offline index, compared with the prior art that only considers search records with a small amount of data to recommend search words for users, it can improve the user recommendation Personalization and targeting of search terms.

Description

technical field [0001] The present invention relates to the technical field of computer applications, in particular to a search word recommendation method and device. Background technique [0002] When browsing video websites and shopping websites, users often input search words to find corresponding video resources or commodities, where the search words may be video names, commodity names, and the like. As for the website, reasonably recommending search terms to users has a very positive effect. For example, if the recent episode "Fengshen Yanyi" is more popular, then when the user logs in to the video website, the video website can directly push the search term "Fengshen Yanyi" to the user, so that if the user is interested in the search term , you can directly click the search button to search based on "Fengshen Yanyi". It can be seen that the way of pushing search terms for users can promote users' consumption of relevant recommended content. [0003] However, in the e...

AI Technical Summary

Problems solved by technology

[0003] However, in the existing search term recommendation methods, only the popularity of the search term itself and the user's previous search records are considered. For example, if a user has recently searched for "Lord of the Rings 1", then the video website is likely to push Search term "lord of the rings 2"

[0004] The data considered in this recommendation method is relatively simple and one-sided, so the personalized and targeted search terms recommended for users are not high.

Images