A Collaborative Filtering Recommendation Method Introducing Video Popularity and User Interest Changes

Abstract

The invention discloses a collaborative filtering recommendation method that introduces changes in video popularity and user interest. The user-video binary correlation matrix is ​​obtained by collecting user behavior data for processing, and the video popularity weight and user interest weight are obtained based on the matrix, and Introduce it into the user similarity calculation process; then find the top K neighbors with the largest similarity with the target user, and predict the target user's interest value in the video that does not produce effective behavior through the similarity with the neighbor user; finally select the largest interest value N videos form a recommendation list to make personalized recommendations for users. The invention fully considers the different popularity of videos in the system and the characteristics of user interests changing over time, which is more in line with objective facts, makes the calculation of user similarity more accurate, improves the quality of collaborative filtering recommendation, and provides video users with video content that adapts to user interests. personalized video recommendations.

Description

technical field [0001] The invention belongs to the technical field of rich media personalized recommendation, and more specifically relates to a collaborative filtering recommendation method that introduces changes in video popularity and user interest. Background technique [0002] With the development of the Internet and its applications, the Internet has ushered in the era of "information explosion". The search technology represented by Google and Baidu can no longer meet the needs of users. First, users may not find the information they want through search engines. , Second, users may not be able to actively express their needs into appropriate words to make the search engine work. As a result, the recommendation system came into being, which can collect the user's historical behavior and feedback information, find resources that meet the user's interests based on this information, and then make personalized recommendations for the user. [0003] Recommendation technol...

AI Technical Summary

Problems solved by technology

The second is to ignore the change of user interest over time

Because the behavior of different users watching video programs in a small time range can better reflect the similarity of users than the behavior of a large time span, so the traditional similarity calculation method does not consider the change of user interest, which is not in line with objective facts.

Images