首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0817105-150420


URN etd-0817105-150420 Statistics This thesis had been viewed 1679 times. Download 21 times.
Author Ying-Na Chang
Author's Email Address No Public.
Department Computer Science and Enginerring
Year 2004 Semester 2
Degree Master Type of Document Master's Thesis
Language English Page Count 59
Title The Design of Personalization and Recommendation Services on Mobile Network
Keyword
  • value-add services
  • recommendation system
  • data mining
  • clustering
  • association analysis
  • Apriori algorithm
  • Apriori algorithm
  • association analysis
  • clustering
  • data mining
  • recommendation system
  • value-add services
  • Abstract The mobile services in this thesis are meant to be the downloadable services via mobile phones, such as games, wallpapers and ring tones. The universality and portability lead us to study the topic about personalization mobile services. Mobile phones can not only provide the interaction between two persons but also show the characteristics of the multimedia. They also perform strong personalization features and have no limitation of the space. Thus, mobile phones can provide good functionalities and information about what users need.
    Due to the restriction of mobile phones, for example, the lower computing power, the smaller screen size, and the insufficient memory, it gives us the motivation of how to provide the proper mobile services from these limitations. Furthermore, the bandwidth and the charging mode are the big issues during the communication in the mobile network, which means that it takes time and money when users search and download services. Therefore, it is important to design a personal service to different kinds of users.
    In order to perform efficiently, most personalized systems ask users to grade the services during the browsing process. This is called an explicit way. But repeated grading process is inconvenient for users. Some scholars proposed an implicit way that analyzed users browsing logs to get their interests.
    The thesis proposes a personalized recommendation system in mobile network. It is a personalized services based on information filtering. By watching the interaction among users, it learns their interests and behaviors and then predicts what kind of the mobile services users need. The system records how much time users spend in it, as well as the frequency users spend in each services. We use Apriori algorithm, association analysis, classification and prediction to cluster services and recommend them to users. By clustering users’ behavior, we can recommend the related services to the users in the same cluster. Not only the theoretical work is presented, but also detailed simulation results are given to verify the applicability of the proposed system.
    Advisor Committee
  • Yo-ping Huang - advisor
  • none - co-chair
  • none - co-chair
  • Files indicate in-campus access at one year and off-campus not accessible
    Date of Defense 2005-07-25 Date of Submission 2005-08-17


    Browse | Search All Available ETDs