Research On The Application Of Mobile Electronic Commerce Information Push Based On BP Algorithm

As the rapid development of the Internet,mobile devices and other information technology and service industry,the current e-commerce is developing in a new direction,which is characterized by mobility and individuality,is gradually known by the public.Mobile e-commerce because of its convenient and efficient way of trading,as well as the convenience of the purchase of goods,has been more and more people's favorite.Although the rapid development of mobile e-commerce has brought us a lot of convenience,but the variety and number of goods are also increasingly diverse,site structure is increasingly complex,which makes it difficult for users to find the information they want in the short time.The development of mobile e-commerce recommendation system brings unprecedented opportunity for its industry.A successful referral system is like another self,but it plays the role of the service,it can be very clear about the user's own preferences and what the user wants,which can improve customer satisfaction and loyalty to the site,thereby improving business performance.On the basis of the above background,this paper mainly does the following work.Firstly,by analyzing the characteristics and nature of the electronic commerce,this paper makes a contrastive analysis of the data mining techniques and recommendation techniques commonly used in the e-commerce recommender system,and then the existing research results based on data mining algorithm.To improve the inherent defects of BP(Error back propagation algorithm)algorithm for data mining method,the main idea is to optimize the BP algorithm by improving the fitness function of GA(genetic algorithm),and describe the core idea and algorithm flow of the new model.And then the experimental test and performance analysis are carried out on this basis.In addition,based on the analysis of the basic architecture of the mobile Electric power recommender system,this paper proposes a Cross-Platform Mobile Commerce Multi-Recommendation System model(CP-MEMRS),and gives the basic framework of the model.The difficulties in practical application of the model are analyzed.