Video recommendation method and server

Abstract

Embodiments of the invention disclose a video recommendation method and a server, the accuracy of video recommendation can be improved and the user experience can be improved. The method provided by the embodiment of the invention comprises the steps of obtaining a group establishment reason for a social group in which a user joins, wherein group types corresponding to social groups with different group establishment reasons are different; according to the group establishment reason, obtaining feature data corresponding to the group establishment reason; for different types of social groups, determining weight values of the social groups according to the feature data; setting the social group with the weight value greater than or equal to a preset value as a candidate group; and recommending a video associated with the candidate group to the user.

Description

technical field [0001] The invention relates to the field of network technology, in particular to a video recommendation method and server. Background technique [0002] With the advancement of multimedia technology and the expansion of social network platforms, video resources are becoming more and more colorful, and related video recommendation methods are also diverse. [0003] Existing recommendation schemes include non-personalized recommendation and personalized recommendation. Among them, non-personalized recommendation generally recommends to users popular videos based on popularity or user demographic attributes such as regions. However, this transitional recommendation scheme The recommendation form is single, and the user experience is poor; personalized recommendation is generally based on the user's historical viewing records, recommending videos related to the viewing records, for example, the same genre, the same director, etc. Referral Programs Referral sour...

AI Technical Summary

Problems solved by technology

[0003] Existing recommendation schemes include non-personalized recommendation and personalized recommendation. Among them, non-personalized recommendation generally recommends to users popular videos based on popularity or user demographic attributes such as regions. However, this transitional recommendation scheme The recommendation form is single, and the user experience is poor; personalized recommendation is generally based on the user's historical viewing records, recommending videos related to the viewing records, for example, the same genre, the same director, etc. Referral program Limited referral sources

The above problems will reduce user experience and affect user stickiness and long-term development of video services

Images