Personalized Classication for Keyword-based Category Proles ?

Personalized classication refers to allowing users to dene their own categories and automating the assignment of documents to these categories. In this paper, we examine the use of keywords to dene personalized categories and propose the use of Support Vector Machine (SVM) to perform personalized classication. Two scenarios have been investigated. The rst assumes that the personalized categories are de- ned in a at category space. The second assumes that each personalized category is dened within a pre-dened general category that provides a more specic context for the personalized category. The training docu- ments for personalized categories are obtained from a training document pool using a search engine and a set of keywords. Our experiments have delivered better classication results using the second scenario. We also conclude that the number of keywords used can be very small and in- creasing them does not always lead to better classication performance.