Colloquium Details
Constructing an Anti Spam-mail System using Fuzzy Inference and Ontology Techniques
| Author: | Jongwan Kim Daegu University, Korea |
|---|---|
| Date: | April 13, 2006 |
| Time: | 15:30 |
| Location: | 220 Deschutes |
| Host: | Dejing Dou |
Abstract
In this talk, I will present a method to construct an Anti Spam-mail System using Fuzzy Inference and Ontology Techniques. This work is based on my previous research and current work with Prof. Dejing Dou. First, the previous work on email filtering task will be given. We proposed a feature selection method by fuzzy inference and its application to spam-mail filtering. The proposed fuzzy inference method outperformed information gain and chi squared test methods as a feature selection method in terms of error rate. In the case of junk mails, since the mail body has little text information, it provides insufficient hints to distinguish spam mails from legitimate ones. To address this problem, we followed hyperlinks contained in the email body, fetched contents of a remote web page, and extracted hints from both original email body and fetched web pages. A two-phase approach was applied to filter spam mails in which definite hint was used first, and then less definite textual information was used. In our experiment, the proposed two-phase method achieved an improvement of recall on the average over the first phase or the second phase only works. Second, I will give an overview of current work with Prof. Dou and his student, Dongwhi. Then this talk is concluded with our future plan.
Biography
Dr. Jong Wan Kim received the BS, the MS, and the PhD degree in Department of Computer Engineering from Seoul National University, Korea, in 1987, 1989, and 1994, respectively. He has been with Daegu University since 1995 and is currently an associate professor. From 1999-2000, he was a visiting scholar at Computer Science Department in UMass at Amherst, working on multi-agent systems. He has written several papers in the areas of information filtering, feature selection, and fuzzy systems. His current research areas include artificial intelligence and data mining. He served as the director of Information and Communication Center at Daegu University during last year. Currently, he has been worked on the construction of anti-spam system using fuzzy inference and ontology techniques as a visiting associate professor at the CIS Department at the University of Oregon.