Research And System Implementation Of Collaborative Filtering Video Recommendation Algorithm Based On Dimensionality Reduction And Clustering

The growth of the Internet has led to an increase in the number of Internet users and information.There is a wealth of useless information that fills people's busy work and lives.People need to spend some time filtering out the useless messages in order to find what they want to see in this huge variety of information.This problem also exists in some film and TV systems,where the number and variety of movies makes it impossible for people to quickly find the movie they want to watch,it takes constant filtering to find the content you want to see.As a recommendation system,the algorithm is the key core.This paper first explains the logic and application of collaborative filtering method.Then,optimization of traditional collaborative filtering algorithms using K-mean clustering algorithm and SVD algorithm for dimensionality reduction,the method of this paper is submitted.The RMSE of the method in this paper is compared with other method through experiments.In order to validate the methodology of this paper,several metrics of the system were also analyzed.Both conclude that the performance of the algorithm in this paper is better.Finally,this paper also studies the workflow of traditional recommendation systems,and combines the current development of recommendation systems,applies this paper's algorithm to the actual development of film and television recommendation system business scenarios,develops a film and television recommendation prototype system,and tests the practicality and scalability of this paper's algorithm.