Research On Personalized Movie Recommendation Algorithm Based On Deep Learning

With the arrival of 5G era,the transmission speed of movies is faster and faster.At the same time,people’s demand for watching movies is more and more abundant,and the software of movies will continue to increase.Today is the era of big data,on the one hand,the amount of data of movie resources is increasing,on the other hand,it is more and more difficult to provide users with accurate demand for movie recommendation.The continuous development of recommendation algorithm can effectively solve this problem.The recommendation algorithm uses data mining technology to extract the historical behavior characteristics of users,and use these characteristics to recommend favorite movies to users.Therefore,it has a good prospect to apply the recommendation algorithm to movie software and provide accurate recommendation service for users..In this paper,a personalized recommendation algorithm based on deep learning is proposed according to the high sparsity of users’ ratings for movies.Firstly,it introduces the historical development and definition of the movie recommendation system,and expounds the basic overview of the recommendation system and the introduction of the recommendation algorithm,with emphasis on the application of deep learning in the recommendation field.Then,a recommendation model Conv FNN based on deep learning technology is introduced,and the components of the model are introduced in detail.Based on the model,this paper optimizes the CNN network embedding layer of the model and introduces glove word embedding model.At the same time,the model is fused with the traditional machine learning algorithm KNN,and a new network model Conv FNN-KNN is proposed.In this paper,we use Movielens,a common open data set in the recommended field,and use the crawler technology to crawl the features of movie profiles to improve the data set.Finally,the data set after processing is used for experiment.Through comparative analysis of experimental results,it is found that the experiment is in line with expectations,showing a positive trend.That is to say,with the increase of the recommended movie sequence,the improved Conv FNN-KNN personalized recommendation algorithm not only ensures the accuracy,but also improves the diversityand coverage of the recommended sequence compared with the traditional PMF algorithm and the deep learning model ConvFNN.