Email Task Management: An Iterative Relational Learning Approach

Today’s email clients were designed for yesterday’s email. Originally, email was merely a communication medium. Today, people engage in a variety of complex behaviours using email, such as project management, collaboration, meeting scheduling, to-do tracking, etc. Our goal is to develop automated techniques to help people manage complex activities or tasks in email. The central challenge is that most activities are distributed over multiple messages, yet email clients allow users to manipulate just isolated messages. We describe machine learning approaches to identifying tasks and relations between individual messages in a task (i.e., finding cause-response links between emails) and for semantic message analysis (i.e., extracting metadata about how messages within a task relate to the task progress). Our key innovation compared to related work is that we exploit the relational structure of these two problems. Instead of attacking them separately, in our synergistic iterative appr...

• Type: Conference
• Year: 2005
• Where: CEAS
• Authors: Rinat Khoussainov, Nicholas Kushmerick
• Keywords: CEAS 2005, Email Clients, Today’s Email Clients, Yesterday’s Email