Design And Implementation Of Film Recommendation Platform Based On Deep Learning

With the development of the current society,the increase in the amount of information has far exceeded the range that people can accept.The simple information retrieval module before can no longer meet the daily needs of normal people.Therefore,a more intelligent recommendation system will naturally come into being.gave birth.In recent years,deep learning has achieved good results in various fields,and many scholars have tried to apply deep learning techniques to recommendation systems,but the current applications are still relatively few,especially for platforms in film and television recommendation.Although there are some researches,it is not thorough.Therefore,this article proposes and builds a high-concurrency and high-availability platform based on deep learning recommendation algorithms for this area of ? ? film and television recommendation recommendation.In today's big data environment,the existing recommendation systems that use information retrieval or information filtering technologies can no longer meet the needs of fast and real-time processing in the big data environment,which hinders the development of recommendation systems.This thesis takes the recommendation algorithm based on deep learning as the research object combined with Hadoop technology,builds a recommendation system with a distributed architecture on these foundations,tests and optimizes the recommendation algorithm based on deep learning,and achieves the expected set from the test results.Availability indicators under.The specific research content of the paper is as follows:(1)Based on Hadoop distributed technology,a high-concurrency film and television recommendation platform was built,and the Hadoop architecture was applied to the film and television recommendation platform,and some improvements were made to solve the problems of user changes and data cold start in the traditional recommendation platform.Designed and implemented a film and television recommendation platform based on Hadoop.(2)In response to the requirements of accuracy,personalization and efficiency in the recommendation algorithm of deep learning,a recommendation model based on deep learning is constructed.According to basic information such as actors,ratings,covers and types of film and television dramas,the recommendation algorithm of deep learning is adopted The trained model can get a movie suitable for users to watch.On this basis,a recommendation algorithm based on deep learning is established.(3)Using key technologies such as Vue,Spring Boot,and Hadoop,design and implement a complete film and television recommendation platform.Finally,the system is tested,including the functionality,load performance,and accuracy of the platform.The test results show that the system has good practicability.