| As an extendible,lightweight,multipurpose metadata description and syndication format,RSS is widely used in web due to its simplicity,standardization,and user-friendly.Until now,how to enable subscribers to grab information rapidly and precisely is a significant research orientation of personalized information service.Additionally,increasing numbers of information spread in web by the methods of RSS,for instance,library,blogs and many media websites(such as BBC,CNET,CNN,Yahoo,Microsoft,etc.)have provided the service based on RSS.Therefore,the personalized information service based on RSS has became a hotspot.With the promulgation of Mobile Intent,getting info from the mobile device is the main approach to gain new information.Building the personalized information service platform for mobile device will provide users with a quick and easy way to capture information.In order to build a Personal Information Recommend System,which offers the service based on RSS,this paper mainly focuses on:(1)Design and empolder an interaction to achieve the goal of the Human-Computer interaction between the users and systems and enable the users to gain info service quickly and easily.(2)Compare the popular XML parsing technology such as DOM SAX etc.,and balance the application platforms and the problems of efficiency of parsing,and finally employ the XML PULL to complete the parsing to XML documents,and put the analyzing results into the databases.(3)Use the Chinese word segmentation system NLPIR(provided by Institute of Computing Technology,Chinese Academy of Sciences)to process the behavior data and get the results of Chinese word segmentation,then analyze the results,and screen out the key words that users interested in from the behavior data to build data model,and update the model according to the behavior databases.(4)Recommend to the users according to the user interest model,and decide whether to recommend this information by the key words that users interested in. |
PDF Full Text Request