Design And Implementation Of Personalized Recommendation System On Electronic Business Platform

With the fast development of the Internet,online-shopping has also become very popular;and now the number of platforms of online-shopping are increasing,especially the Taobao C2C model which appears to greatly contribute to the development of domestic online-shopping.Online-shopping achieves another way of retail.This approach gets away with the problems of the traditional retail model with stores,and effectively solves the problems of the long-tail goods in retail and product demonstration.With online-shopping,customers will be able to save their time which is beneficial.The online shopping however will also have situations with information overload.Platforms will upload and demonstrate all the product information,and the costumers can process the overloaded information by using searching engine.But the customers cannot tackle the problem of information overload,in the absence of clear targets.How to use the recommending system for the costumers to overcome the information overload has become an urgent issue to be solved by the electronic business platforms.In this context,this paper puts forward the design and implementation of personalized recommender system based on electronic business platform.Through the personalized recommendation,it can effectively improve the user's ability to deal with commodity information,thereby increase the purchase rate of the entire platform.The main contents of this paper are as follows: A detailed requirement analysis of the system is carried out.The main requirements are analyzed from the aspects of system construction goal,system feasibility analysis,functional requirement analysis and system non-functional requirements.The functional requirement analysis is mainly from two aspects: the overall demand analysis of the system and the functional requirement analysis of the system.The system is designed and implemented on the basis of requirement analysis.System design mainly includes system architecture design and system module design.The system module mainly includes data processing module,personalized recommendation algorithm module,recommendation result processing module,recommendation request processing module,commodity management module and user management module.Based on the system design,the system is implemented.In this paper,Hadoop is used to store and log the ETL.The ETL process of the log is analyzed by hive tool,and finally the data is stored onHDFS.Then,it implements the personalized recommendation algorithm module,which mainly implements user-based filtering and object-based filtering in the collaborative filtering algorithm.Secondly,the results of the personalized recommendation algorithm module are processed,including filtering and ranking processing,so as to improve the novelty of personalized recommendation list.Finally,the system needs to obtain the recommended list of users from the recommended system through the http request.