The prediction of internal human states—what a person is thinking and feeling—by computers remains an elusive yet attractive goal. The potential applications of such a system are many, including but not limited to lie detection, consumer satisfaction and interest, autonomous vehicle safety, enhanced social networking, and even robotic friendship and companionship.
We believe that scientific progress towards such lofty and exciting applications as described above has been slowed because prior efforts to create an autonomous computer system capable of predicting what a human being is thinking or feeling have been largely predicting or detecting “emotions” in humans, rather than detecting more empirically falsifiable events. There is much scholarly disagreement as to the exact nature of what emotions actually are, and thus Affective Computing dealing with emotional prediction is forced to rely on poorly-defined and often-contradictory ground truths.
In our research, we focus on predicting discrete events based on the expressive behavior of human subjects. Our systems use a user-dependent method of analysis and rely heavily on contextual information to make predictions about the meaning behind subject expressive behavior. Our system’s accuracy and therefore usefulness are built on provable ground truths that prohibit the drawing of inaccurate conclusions that other systems could too easily make.
While the current application of our research idea—a Contextually Informed Program for Hyper-personalized Empathetic Report, or CIPHER—is modest (predicting player behavior during a game of poker), we believe it to be the foundation upon which much more extensive empathetic and affective analysis systems could be solidly and eventually built.
|Commitee:||Leeds, Daniel, Zhao, Yijun|
|Department:||Computer and Information Science|
|School Location:||United States -- New York|
|Source:||MAI 58/04M(E), Masters Abstracts International|
|Subjects:||Psychology, Artificial intelligence, Computer science|
|Keywords:||Affective computing, Computer vision, Contextual awareness, Event prediction, Machine learning, Poker|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be