Study On Mobile Client Personalized Push Based On Context Awareness

Mobile internet has sprung up and developed at a rapid rate that no forecaster has expected. Therefore, those who can foresee the future of mobile internet will be able to create more economic miracles. With a huge user group, mobile internet now holds the upper hand in the Internet sector. The significant improvement in the speed of Internet access via mobile phones enables mobile clients to enter a broader range of application contexts in people's life, such as shopping, housing and transportation. Moreover, the "Internet+" strategy ignites explosive growth of mobile internet and also drives development of other industries. However, conventional shopping applications (APPs) only passively provide massive product information and often fail to display products the users want to buy, preventing them from quickly selecting satisfactory products according to their preference. A mobile APP's failure to provide considerate service will cause a loss of users in the long term.In order to solve this problem, this paper provides an analysis of the concepts of context awareness and personalized push as well as a review of previous studies. A method of personalized push based on context awareness was developed for mobile clients based on a discussion about the characteristics of contextual information, common personalized push methods and their advantages and disadvantages. This method uses explicit acquisition and implicit acquisition to collect contextual data of APP users. A user identification method was proposed to allow APPs to identify unregistered users and push personalized messages to them. Then, a user preference model was constructed using a modified user-item evaluation model, yielding an expression for the degree of preference for a product, denoted Pug. The Apriori algorithm, a classical algorithm for recommending products based on association rules by was modified. Temporary table was introduced so that the invalid data items can be cleared when calculating support and scanning and comparing transaction data. This can cut the number of counted tables in each traversal, improve the efficiency of the algorithm, reduce the pressure on the servers, and thereby enhance the performance of personalized push systems.A personalized push system based on context awareness was designed. This system can perform acquisition, processing, modeling and mining of users'contextual information. Shopping APPs with this system can provide considerate service that users expect through personalized push service and customized push information page. By forecasting goods the users may be interested in, this system can release users from information overload. Personalized, accurate recommendation of products can greatly increase sales volumes, enhance brand reputation, and improve user experience, and as a result, the APPs will gain more loyal APP users and become more competitive in the fierce market competition.