Design And Implementation Of Movie Recommendation System Based On NTM Neural Network

With the rapid development of Internet technology,people's dependence on the Internet has become more and more obvious,and it has been scattered in people's daily lives.In order to add a little fun to people's boring and boring life,the major video websites provide a huge number and a wide variety of movie resources for users to watch,so each time they watch a movie,users need to spend a lot of time to pick Movies that match your preferences.In order to solve this problem,researchers have applied personalized recommendation technology to the movie and video industry.The results show that the technology can effectively alleviate the problem of information overload,thereby improving the user's experience and increasing the user experience.Use stickiness to the system.At present,deep learning technology is developing in full swing and has achieved outstanding results in many fields.Therefore,the use of deep learning methods to carry out research on personalized movie recommendation systems has important practical significance and use value.First,this article briefly explains some traditional recommendation algorithms and their respective advantages and disadvantages;second,based on the analysis of historical user behavior data in the movie recommendation system,and in order to make the recommendation results more in line with user preferences,In this paper,social networks are introduced to build a mathematical model;again,considering that the accuracy of the rating prediction will also affect the overall performance of the movie recommendation system,an auto-encoder is introduced,which is to re-use the data originally used for movie ratings Construction,so as to get a more accurate movie score;finally,using the movie data of Douban website as a data set,the traditional project-based collaborative filtering algorithm,the recommendation algorithm based on Markov chain and the recommendation algorithm based on NTM neural network The experimental comparison shows that the recommendation algorithm based on NTM neural network has better effect on accuracy and average response time.Using the NTM neural network with better experimental results as the recommendation algorithm,a movie recommendation system based on the NTM neural network was designed and completed.Finally,the function and performance of the movie recommendation system was tested to verify that the system can indeed provide users Movies that are more personalized and more in line with your preferences.