TITLE: Automatic Induction of Rules for e-mail Classification

PRESENTER: Eric C McCreath Basser Department of Computer Science, University of Sydney

DATE: Friday, 4th May 2001 TIME: 12:00 - 13:00

PLACE: CSE K17 1st Floor Seminar Room

ABSTRACT:

Many users receive large amounts of email. Since a substantial part of that mail is kept for future reference, it is unsurprising that many mail tools allow users to create filtering rules that automatically do actions like saving mail in suitable folders. Unfortunately, most users do not make much use of this facility.

Our current research focuses on building a system to assist users in managing electronic mail. It will do this by learning rules for classifying email so that it can assist the user in filing messages.

This seminar will review different learning approaches used in addressing this problem. These include: Keyword spotting, Naive Bays, and TF-IDF approaches. We will also look at different ways these have been incorporated into email clients. A system has been developed to demonstrate how an ILP strategy may be employed to assist users in managing electronic mail. This system will be demonstrated. Finally, we will discuss a variety of challenges this problem poses the machine learning community.

Note that, this project involves a team from The University of Sydney consisting of Elisabeth Crawford, Judy Kay, and Eric McCreath.

BIOGRAPHY OF SPEAKER:

Eric completed a BE in computer engineering and The University of New South Wales(UNSW) and then went onto completed a PHD at UNSW under the supervision of Prof Arun Sharma. Ericês research field is machine learning, while concentrating in the specific area of inductive logic programming (ILP). Currently Eric is working as a Lecturer at The University of Sydney.

Seminar Convenor:

Bernhard Hengst
Tel: +61-2-9385-3988,
E-mail: bernhardh@cse.unsw.edu.au
School of Computer Science & Engineering, UNSW.