Search recommendation method and recommendation device, electronic equipment and storage medium

Abstract

The invention relates to a search recommendation method and device, electronic equipment and a storage medium, and the search recommendation method comprises the steps that candidate search results are obtained based on search terms; obtaining a first search result and a second search result from the candidate search results; performing generalization processing on the features of the first search result to obtain generalization features of the first search result; the generalization features of the first search result are input into a behavior estimation model, behavior estimation parameters are obtained, and the behavior estimation parameters are used for evaluating the probability that the first search result is subjected to a preset operation behavior by the target account; and based on the behavior estimation parameters and the historical behavior scoring probability value of the second search result, determining a recommendation sorting result of the candidate search results. According to the method, all output search results can be reasonably ranked, the Mattai effect is reduced, the recommendation ranking position of the new search results can be set in the front, a user can continuously check the new search results, and the search recommendation accuracy is improved.

Description

technical field [0001] The present disclosure relates to the technical field of data processing, and in particular to a search and recommendation method, a recommendation device, electronic equipment, and a storage medium. Background technique [0002] In related technologies, in the scene of video search, a situation often occurs: a newly uploaded video by a user or a video newly added to the index system, although the user may "define" some tags or text information for this video, but Because the "new work" lacks the interaction data between the user and this video, when sorting it, it will be compared with the "old" (the user interacts with the video more times, for example, the number of retweets and the number of likes are high) , Watching time is long enough, etc.) Videos appear to be "disadvantaged" in recommendation, resulting in a lower ranking of new video recommendations. [0003] For example: under the same search term, 10 videos are indexed, 9 videos belong to ...

AI Technical Summary

Problems solved by technology

[0004] The present disclosure provides a search recommendation method, recommendation device, electronic equipment, and storage medium to at least solve the problem in the related art that video search results tend to rank new search results at the end, resulting in low accuracy of search recommendation

Images