Intelligent E-mail Management System

As e-mail becomes more popular over networks, the problem of receiving a lot of undesired e-mail arises. All of the previous systems developed to solve this problem may be divided into two different categories: client-side and server-side, from the view of the location of the filter. In the former case, after all e-mails are downloaded, the client-side e-mail system can filter or process the undesired e-mails according to the user's e-mail reading preference. This way always causes a waste of network traffic. In the latter case, all e-mails may be first filtered in the server according to some general and predefined filtering rules, and then the remaining e-mails are downloaded to the client. It results that all users may only share a small predefined set of filtering rules. We propose a new architecture of an e-mail management system, called the Intelligent E-mail Management System (IEMS), integrating client-side and server-side to help users manage their e-mails. A mining agent is used in this system to find rules from the user's behavior for reading e-mail automatically. The prediction of the e-mail is made before transferring it to the client, therefore, the undesired e-mails can be filtered at the server to save network traffic. In addition, the feedback mechanism is provided to refine the prediction accuracy based on each user's decisions for reading e-mails. Experimental results show that IEMS can infer e-mail reading preference of individual users correctly up to the accuracy about 80%. We may conclude that IEMS can help users manage their e-mails effectively.