AI in Video Recommendation System

Abstract

Short videos are very popular all over the world. Video recommendation system is an essential part in it. It can help people to watch the video that they are interested in. This paper is written for study the specific principle of the video recommendation system. The result was getting through relative literatures and actual test. Short video recommendation systems typically use collaborative filtering and deep learning techniques to achieve this. Collaborative filtering comes in two types: user-based and content-based. User-based collaborative filtering recommends videos to new users based on the viewing behavior of similar users. Content-based collaborative filtering uses video features and similarity to recommend similar videos. Finally, this paper shows how the video web set can learn what is the user’s interest.