Professor Michael Terry will be presenting a talk on Thursday, November 20th from 11:00 am until 12:30 pm in room BA5187.

Please join us in welcoming Professor Terry!

“Interactive Systems Need to Know How to Read the Web and Watch YouTube”

In this talk, Professor Michael Terry will argue that there is great value in interactive systems that can learn how to accomplish tasks by “reading” web-based tutorials and “watching” how-to videos. He will focus primarily on text-based documents and search queries, and show how techniques from the fields of machine learning and information retrieval can be leveraged to extract streams of “how-to” information from web-based resources and instrumentation logs. These information sources enable a new class of interactive system that is more aware of the tasks it can perform, as well as how to accomplish these tasks. Importantly, this awareness continually evolves and tracks how the user community actually uses the system.

Biography Michael Terry is an associate professor in the Cheriton School of Computer Science at the University of Waterloo, where he co-directs the HCI Lab. His research lies at the intersection of HCI, machine learning, and information retrieval. His current projects include machine understanding of instructional materials, task-centric user interfaces, and interactive machine learning systems designed to assist the digitization and cataloging of millions of biological specimens in London’s Natural History Museum.