Computing and Information Systems - Theses

Permanent URI for this collection

Now showing 1 - 1 of 1

Predicting the priority of email messages using stylistic features

Ward, Belinda ( 2001)

LASSIE (Learning Apprentice System for Sorting Incoming Email) is an email management tool that has been constructed in order to demonstrate a possible solution to the problem of email overload. It is designed to help users manage their incoming messages more efficiently by sorting the messages according to their level of urgency. This thesis discusses the motivation and requirements for such a system, and describes the design, implementation, and evaluation of LASSIE. We identify stylistic features that are good indicators of message urgency, describe how these features can be extracted from plain text messages, and present an algorithm for inferring the relationship between these attributes and the message urgency for any individual user. We discuss the considerations involved in constructing a practical email management tool and describe the design decisions that were made in constructing LASSIE. The results of a user trial indicate that LASSIE is able to achieve a high level of predictive accuracy and is readily adopted by users, although improvements in its long-term performance are desirable.