Research On The Recommendation Method Of Boku Book Mall Based On User Data Mining

With the rapid growth of the volume of data in the field of electronic commerce,electronic commerce has entered the data based era,enterprises hope to seize this opportunity,the data will be used effectively,personalized recommendation has become important research content of the current electricity supplier data application,its role is to provide users with high quality personalized service,a large number of users through mining data hiding information to determine the user preferences,and recommend personalized merchandise to them,promote the user's shopping behavior,so as to improve sales,get more profits.In addition,personalized recommendation but also to tap the potential of the user's shopping intentions,to recommend to customers so far has not been exposed to the possible purchase of goods information.Not only can bring quality services to users,but also for electricity providers to bring unprecedented profits,which is one of the reasons for the wide application of personalized recommendation.Therefore,to improve the level of e-commerce personalized recommendation can improve the service quality of business enterprises.In the mall Boku book as the research object,analysis of the type of data mart produced,characteristics and the existing problems in the application of user data,and then analyzes the process of data mining and mall recommendation method,the user data is divided into text comment data,the basic characteristics of the data,transaction data,text data mining in the comments,will be false comment filtering,book attribute extraction based on emotional words,construction of Library and attribute database;in mining user basic characteristics and the characteristics of trading behavior in mining properties are derived,using FCM clustering method to divide users into high value users,medium value users,ordinary users value,low value users,in four classes in using improved user similar user similarity calculation method;in the recommendation process,the user Book score combined with data from the book attribute text comments Into the user attribute value similarity score,filling user attribute value sparse matrix using the nearest neighbor and then get the target user,the user preference of attribute value to calculate the user set;not rated scores,the target user attributes on the value of all goods mean score ranking,the formation of the user's Top-N recommendation list.To complete the recommendation,the accuracy of the proposed method is proved by experiments.Through the design of user data mining and Boku Book Mall recommendation method,a full use of user data store generated,with the recommended collaborative filtering theory,enrich the theory and application of data mining;on the other hand to design personalized recommendation method Boku Book Mall precise recommend books to users,targeted recommendation method for the Boku book mall design to help online book shopping mall business.