Design And Implementation Of Film And Television Recommendation System Based On Cooperative Filtering Algorithm With Coupling Similarity

In recent years,with the widespread use of the Internet,the data resources it carries are increasing exponentially.It has gradually developed from the era of lack of information in the early stage to the era of information overload.The accompanying online film and television has also become rich,and its market prospect has become more and more popular,and it has become more and more imaginative space.However,with these changes,when a single user looks for the information and resources he needs in the platform,his energy and time will become more and more.Therefore,it has become a rigid demand that online film and television platform must realize to help users find the resources efficiently and quickly,and its effect determines the success of the platform.Personalized recommendation system as an efficient information filtering means,excellent personalized recommendation system can meet the needs of users in a short period of time,the realization of the system mostly uses neighborhood based collaborative filtering algorithm,some may also mix content-based recommendation algorithm or demographic based collaborative filtering algorithm,etc.,the above algorithms have been proved by practice,but also achieved Good results have been achieved,but the problems of "cold start" and "sparsity" have not been solved effectively.The main research content of this topic is to build a personalized recommendation system of online film and television platform,through personalized recommendation for users,automatically recommend the film and television resources that may be of interest to users,help users quickly select their favorite film and television resources from the mass of information,so as to save users’ time and make users have a better experience of appreciating film and television works And effectively improve user stickiness.In this paper,several common algorithms are analyzed and compared,and each algorithm’s shortcomings and advantages are found,which are combined with each other to avoid the situation that one algorithm alone can not meet the needs of improving user experience.In this paper,we will study and implement the collaborative filtering algorithm film and television system combined with the similarity of coupling objects,so as to further improve the accuracy of personalized recommendation system,effectively help users to more quickly select interested films and television,and solve the problems of "cold start" and "sparsity" as much as possible.This paper introduces three kinds of filtering algorithms: user based,object-based and model-based collaborative filtering algorithm,and describes the advantages and disadvantages of the above algorithm and the use scenarios.At the same time,it also designs and verifies the similarity algorithm of coupling objects.Finally,on the basis of the above algorithm,the requirements of personalized recommendation system for online film and television platform are investigated,and on this basis,the overall framework,business module,business process and database are designed,so as to realize a personalized recommendation system.Finally,the system is tested,and the expected good results are achieved.