The Design And Implementation Of An Explainable Movie Recommendation System Based On Knowledge Graph

With the rapid development of network information technology,the amount of data in the Internet is increasing day by day,and the content also tends to be homogenized.In order to solve the long-tail effect of content,recommender systems emerge as the times require.At present,most recommendation systems generate recommendation results based on implicit information such as historical records.When users first contact the system,their trust and dependence on the recommendation results are very low.Therefore,the method of recommendation explanation is used to increase the communication between the system and users and strengthen the recommendation system.transparency and user trust.The explainable recommendation system based on knowledge graph adopts the C/S architecture mode,and its main architecture includes business presentation layer,logic processing layer and data storage layer.The system takes movies as the recommended object,and its main functions include movie information collection,movie review sentiment analysis,movie recommendation and system management.The movie information collection function is responsible for crawling movie data from movie websites,analyzing emotional information in movie reviews and storing them in a structured manner.The movie recommendation function collects user preferences,finds similar movies from the knowledge graph for recommendation,and generates recommendation explanations according to the search path.The system management function is responsible for managing system permission logs and other related information.The interpretable recommender system designed in this way,while making recommendations for users,uses templates to generate explanations and present them to users,making the recommendations more transparent.The system develops the App side and the server side in terms of implementation.The App is built using the Flutter framework and is compatible with Android and iOS systems.The server is built with the Django framework,and the enhanced management system background is generated through the xadmin model,and the system timing tasks are used for regular recommendation,which makes the recommendation results more diverse.The system function test and performance test show that the system can provide personalized recommendation service stably.