Design And Implementation Of Application System And APP Tags Mining

With wildly prevalent of smartphones, mobile internet is constantly changing theway of people’s life and taking users from PC to mobile terminal. All these changesare based on abundant of mobile Application, which makes mobile terminal mainlyhave all functions of the computer and use more convenient. However, with thecontinually growth of App numbers, the demands of users retrieve are alsoincreasing. Therefore, traditional search engines are difficult to return informationaccurately to users’ need, which has wasted a lot of time.The main content of this article is through text mining to get App concept system,attributes and semantic tags from two aspects which are web information and querylogs. Thus, this article is through these tags to build a vertical search engine in faceof App field, and to conduct the App inquiry and recommendation service.The concept system of App tag mining is achieved by getting information of Appwebsite navigation. Through analyzing the navigation bar to get classifiedinformation of App, and then incorporating all classified information based oncosine similarity and synonym extension classify system fusion strategy which aretwo concept came up in this paper. More specifically, to get attribute tags is to grabthe App description as App text which is through keyword extraction technology;however, Semantic tags are achieved by query log. For the query, title, url and clickbehavior information in query logs, through a series of rules and filtering strategyproposed in this paper to make the query and App correspondence, and then throughthe participles, lexical analysis, syntactic analysis and other methods to deal withthe query. The advantages of the tag mining method above are no need for artificialparticipation and saving a large amount of resource consumption.The whole application system is set up by the Django framework and through theLucene full-text to search toolkit and to construct index for label data, which hasrealized a retrieval system with the function of semantic analysis. After function test, meet people’s requirements of App domain retrieval.