Video recommendation method and system

Abstract

The invention discloses a video recommendation method. The method comprises the steps that user information data, video information data and behavior information data generated when a user watches videos are collected; a factorization machine model is established according to the collected data, and a reinforcement learning network model is trained; historical behavior information data up to the video currently watched by the user is acquired, and a recommended video of the user is obtained based on the factorization machine model and the trained reinforcement learning network model according to the historical behavior information data; and actual feedback information of the user on the recommended video is recorded, and the factorization machine model and the trained reinforcement learning network model are optimized according to the actual feedback information. The invention furthermore discloses a video recommendation system. Through the video recommendation method and system, video recommendation accuracy and speed can be effectively improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a video recommendation method and system. Background technique [0002] With the continuous enrichment and rapid update speed of various video platforms, it is difficult for users to find the content they are really interested in in the face of excessive information. Therefore, we recommend video content that meets their interests and tastes to improve video recommendation. The accuracy rate has become one of the key research areas of major video platform operators. [0003] Currently commonly used recommendation algorithms include recommendations based on content, knowledge, and graphs, and collaborative filtering recommendations. Among them, collaborative filtering recommendation is widely used, including user-based collaborative filtering and item-based collaborative filtering two types of recommendation algorithms. Collaborative filtering can produce higher accurac...

AI Technical Summary

Problems solved by technology

Among them, collaborative filtering recommendation is widely used, including user-based collaborative filtering and item-based collaborative filtering two types of recommendation algorithms. Collaborative filtering can produce higher accuracy in different application scenarios, but when the number of users or the number of items When there are more, the performance of the recommendation system will be limited, and it is difficult to solve the cold start problem

Images