Research On The Application Of Ontology In WAP-based Mobile Search

With the popularization of smart mobile phone and the coming of the3G ages, mobilesearch has been challenged and impacted more and more. The widely use of APP(Applications), leads to the living space of mobile search is getting smaller and smaller.Mobile search only provides universal service, but the APP can provide accurate and preciseinformation. The disadvantage of mobile search is obvious. Mobile search must beintelligent, personalized as soon as possible, in order to have enough living space in thefuture.The greatest charm of mobile search is "real-time on-site", which means you can getthe information you need at any time, any place. Arbitrariness of time and space, resultingin ever-changing search needs of the mobile users. Personalized mobile search mustconsider the different information needs based on the difference of mobile user identities,hobbies, time and context, smart infer the user’s individual needs, and provide personalizedservice.This paper presents an ontology-based personalized search engine to study the user’sinterest correlation, time and context. Firstly, search engine uses implicit uniqueness ofmobile users’ MSISDN (Mobile Station international ISDN number) establish the user’spersonal interest model based on analysis of the preferences of the user’s interest, andanalysis of user queries preferences related to time and context. Query, combined with userpreferences, access time and context, mobile search engine infer the user’s query intention,clear users’ needs, to meet the personalized requirements of mobile users’ search.The contributions of the paper are summarized as following：(1) This paper presents a method of using mobile phone MSISDN and pseudo-code asan identifier to create user ontology. (2) In order to analyze the mobile search "real-time on-site" better, a method ofanalyzing user queries needs is presented in this paper, which based on user ontology,combined with time, context.(3) The concept of access time parameters is presented in this paper. User access timedivided into a plurality of time slots according to the user’s preferences change, the methodof time series analysis was introduced to analyze the relationship between the demand forusers to search and access periods.