Towards supporting the elderly age in place: multimodal reference resolution in the ROBOHELPER project

 Dr. Barbara Di Eugenio
11am Thursday, October 18, 2012
EB3105

 ABSTRACT: The goal of the ROBOHELPER project is to design and develop robotic assistants that could help
the elderly live independently. In the first phase of this project, we are focusing on the multimodal interface required for the human to communicate with the robot.  I will first describe the data collection of human-human interactions we carried out with elderly subjects.  I will then discuss the pronominal and deictic reference resolution model we developed. A novel aspect of our model is that it includes "haptic-ostensive" actions, namely, gestures that  involve manipulating an object. Our results show that the presence of simultaneous gestures improves coreference resolution for personal and deictic pronouns. I will conclude by illustrating current work on recognizing haptic gestures on the basis of the pressure sensor signals we collected.

 BIO: Barbara Di Eugenio is Associate Professor in the Department of Computer Science at the University of Illinois, Chicago campus. There she leads the NLP laboratory (http://nlp.cs.uic.edu/). She obtained her PhD in Computer Science in 1993, from the University of Pennsylvania. She is an NSF  CAREER awardee, and a past treasurer of the North American Chapter of the Association for Computational Linguistics. She is one of the founding and  managing editors of the Journal of Discourse and Dialogue Research, she was program co-chair of INLG 2012, and she is program-co-chair of SigDIAL  2013. Her research has been supported by NSF, ONR, Motorola, Yahoo! and the Qatar Research Foundation.