User Interest Modeling And Personalized Recommendation Study Based On Mobile E-commerce

With the development of the Internet, a lot of people’s traditional way of shopping shift from offline to online, e-commerce has brought to people’s shopping more convenient, affordable experience, online shopping model has become widely accepted and prevalent in the new consumer mode. In the ten years of development, the scale of e-commerce expands gradually, a variety of electronic business platforms, third-party merchants grow rapidly, information overload trends in e-commerce platform appears. How do people find what they need in the vast amounts of information in the field of e-commerce has become an urgent need to solve, the problems is particularly prominent in the arrival of the mobile Internet boom. Due to the particularity of the mobile terminal and the effect of situation to user consumption, mobile e-commerce information overload problem is even more prominent. A good personalized recommendation method allows users to find the precise needs of the project in the flood of information, thereby enhancing user satisfaction and platform stickiness, but also bringing the business more sales profits, realize users, platforms and businesses win-win situation. At the same time, due to the particularity of mobile commerce, in order to better meet customer needs, the research of mobile electronic commerce personalized service recommendation based on situation is essential.Firstly, This article discussed traditional e-commerce personalized service recommendation method on pc terminal, described the basic theory and the overall framework, introduced the commonly used method of personalized recommendation, the effect of personalized recommendation method in e-commerce. Based on this, explained requirements of personalized service recommendation algorithm under mobile e-commerce after summarizing the characteristics of mobile e-commerce. Secondly, in order to better study personalized recommendation method under a mobile environment, the article studied user behavior and modeling under mobile environments, after analyzing the factors that influence the behavior of users to buy under a mobile environment and common user model building method, combined with common two-dimensional modeling method, the user-item matrix user modeling method, proposed "position-user-project" three-dimensional model of the mobile user, user interest SUI model based on mobile e-commerce, each point in the space represents interest of users to a project under any situation. In this model, based on double filtration method recommended by personalized recommendation that the first filter position situation, identify key business situation occurred with the current position of a similar situation in the historical location, this business items as recommended by the library, and then based on this recommendation library user interest filter to find items of interest Suggest library users, according to the degree of interest in these items are sorted from high to low, take the first N items recommended to the user as the final result.