Skip to Content

CSAAW Talk: Considering Human Conversation as a Complex System: Some Early Success and a Lot of Failures

Prof. John Sherry and Qi Hao
Thursday, March 23, 2017
12:00-1:00 PM
438 West Hall Map
Imagine you are having a delightful conversation with a friend on a sunny afternoon in the Quad. With what degree of precision do you think you can predict the next thing your friend will say? Maybe you can guess the type of speech act, but the exact words? What about what your friend will say in five minutes? For that matter, with what degree of precision to you think you can predict exactly what you will say in five minutes? It seems impossible. However, we know conversations must follow some type of agreed upon rules, otherwise we wouldn’t be able to carry on conversions or to produce the types of outcomes that we desire (e.g., agreement, support, idea development). Though conversations are the basic mechanism by which humans accomplish everything from meeting a new person to building an economic system, we know very little about how conversations work. If we understood, Siri, Cortana, Alexa and Watson would be able to form a study group! We will be presenting findings from research in which we are trying to discover the rules that give rise to the apparent complexity of human communication. We’ll describe a novel approach to studying conversations and present a variety of results ranging from some successes to lots of failures, in hopes of stimulating some great feedback from the CSAAW group.
Sign up at:
Building: West Hall
Event Type: Workshop / Seminar
Tags: Electrical Engineering and Computer Science, Information and Technology, Interdisciplinary, Research
Source: Happening @ Michigan from The Center for the Study of Complex Systems