Intelligent Recommendation Platform for Film and Television Based on Machine Learning Algorithms

With the development of Internet science and technology, the target demand of film and television intelligent recommendation algorithm is also changing, and the intelligent algorithm based on machine learning can effectively cope with the current increasingly personalized film and television recommendation. In the whole recommendation system construction, how to improve the accuracy rate of film and television recommendation and reduce the system recommendation time is the key problem that needs to be solved urgently. Based on the analysis of film and television recommendation system and simulation experiments, this paper applies the clustering mining algorithm based on ML, combines the data set samples, similarity and scoring calculation of video software, and discusses the data results to draw the following conclusions. In the context of big data on the Internet, the film and television intelligent recommendation system using clustering mining algorithms has higher system recommendation accuracy than traditional recommendation methods for all dataset samples, with an average improvement of about 6.8%. Meanwhile, in terms of system recommendation time, the clustering mining algorithm is lower than traditional recommendation methods, and the average reduction is 10.8ms. This indicates that the intelligent recommendation platform for film and television based on clustering mining algorithms has higher overall accuracy and efficiency in film and television recommendation. Finally, the intelligent recommendation platform for film and television proposed in this paper provides people who like film and television at present with a choice of works more in line with their preferences, and has certain reference value in practical applications.