Abstract
The online audience for newspapers continues to grow in both numbers and sophistication. However, most online newspapers still offer all users the same content and fail to raise an individual user's satisfaction and to increase revenue for an online newspaper company. The Dynamic Dissemination of Digital Information (DDDI) system developed here provides personalized online content based on a user's preference. The DDDI system is a hybrid application of machine learning techniques, combined with Bayesian content classification, neural content clustering, and online content matching. In addition, the system adopts a time- varying user profiling method (i.e., current-page, last-session, and recent-sessions profiles), which considers dynamic changes in the user needs or preference for digital content. Through the dynamic profiling, this system overcomes the shortcomings of the static user profiles that have been used in the content personalization research so far and creates a one-to-one relationship between the content provider and the users.
| Original language | English |
|---|---|
| Pages (from-to) | 96-111 |
| Number of pages | 16 |
| Journal | Journal of Organizational Computing and Electronic Commerce |
| Volume | 19 |
| Issue number | 2 |
| DOIs | |
| State | Published - Apr 2009 |
Keywords
- Content classification
- Content matching
- Content personalization
- Documents clustering
- Recommender system