Research And Implementation Of User Behavior Prediction Algorithm Based On Improved CBS Algorithm In Smart Home

With the rapid development of science and technology,especially the large-scale commercial use of 4G(fourth generation mobile communication network)technology,5G(fifth generation mobile communication network)technology is expected to be commerci-ally available in 2019,making the interconnection of all things possible.The derived smart home has become a hot topic in the industry.However,the current hot industryresearch mostly from the perspective of the system,dedicated to information technolo-gy and system functional requirements of traditional products.The convenience of op-eration has been greatly improved,but it still needs human control to achieve thecorr-esponding functions.Therefore,there is still much room for improvement in intelligence.Modern smart home systems can collect user behavior information and learn userbehavior by analyzing related information,establish a "behavior pattern collection"forusers,and ultimately achieve predictive user behavior.The purpose is to actively andin advance complete some necessary operations for the user.Smart homes can be converted from passive service provision operated by human to active service provision.In order to achieve the above objectives,this paper proposes a user behavior prediction algorithm based on CBS(Customer-based servicing).In order to verify the performance of the algorithm,the algorithm was coded and embedded into the Java project.The specific work is as follows:1?This paper presents a user behavior prediction algorithm based on improved CBS algorithm.The main flow of the algorithm is as follows:Firstly,the relationship between user behaviors is obtained through association rules and temporal relations.Secondly,the paper describes the time partitioning and intersection and union operations in the CBS algorithm and analyzes the advantages and disadvantages of the CBS algorithm.On this basis,the paper introduces the machine learning technology to solve the influence of the surrounding environment on the user behavior and introduces the "cold heat treatment" idea in the database to solve the problem of time conflict in the multi-behavior situation.2?In order to verify the effectiveness of the algorithm,this paper encodes and embeds the above algorithm into a Java project.The project uses Spring MVC as a model and is mainly divided into four parts:user behavior prediction algorithm,algorithm test,Java project settings,and user information display.Through the project,user behavior prediction algorithm is verified.The verification results show that compared with the CBS algorithm,the improved algorithm has more prediction successes,higher success rate,better stability and effective reduction of user interference.At the same time,the project has features that are independent of the smart home platform,enabling cross-platform operations.