Recommending and evaluating choices in a virtual community of use
CHI '95 Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Fab: content-based, collaborative recommendation
Communications of the ACM
Learning and Revising User Profiles: The Identification ofInteresting Web Sites
Machine Learning - Special issue on multistrategy learning
Modern Information Retrieval
Hi-index | 0.00 |
This study suggests a recommendation agent system that the user can optimally sort out incoming email messages according to category. The system is an effective way to manage ever-increasing email documents. For more accurate classification, the Bayesian learning algorithm using dynamic threshold has been applied. As a solution to the problem of erroneous classification, we suggest the following two approaches: First is the algorithmic approach that improves the accuracy of the classification by using dynamic threshold of the existing Bayesian algorithm. Second is the methodological approach using recommendation agent that the user, not the auto-sort, can make the final decision. In addition, major modules are based on rule filtering components for scalability and reusability.