Ugan Yasavur

Florida International University School of Computing and Information Sciences


Lecture Information:
  • April 24, 2024
  • 12:18 PM
  • ECS: 349

Speaker Bio

Ugan Yasavur is a Ph.D candidate in the School of Computing and
Information Sciences, at FIU. He has been a member of the Affective
Social Computing Laboratory since 2011 where he works under supervision
of Dr. Christine Lisetti. He received his Master degree in Computer
Science from FIU in 2012, and his Bachelor degree in Computer
Engineering from Izmir University of Economics in Turkey. Ugan Yasavur’s
research interests include spoken dialogue systems, dialogue management,
natural language processing, affective computing and health promotion
systems. Ugan has published 12 refereed scientific articles.

Abstract

Research endeavors on spoken dialogue systems in the 1990s and 2000s have led to the deployment of commercial spoken dialogue systems (SDS) in microdomains such as customer service automation, reservation/booking and question answering systems. Recent research in SDS has been focused on the development of applications in different domains (e.g. virtual counseling, personal coaches, social companions) which requires more sophistication than the previous generation of commercial SDS. The focus of this research project is the delivery of behavior change interventions based on the brief intervention counseling style via spoken dialogue systems.

The overall objective of this research is to develop a data-driven, adaptable dialogue system for brief interventions for problematic drinking behavior, based on reinforcement learning methods. The implications of this research project includes, but are not limited to, assessing the feasibility of delivering structured brief health interventions with a data-driven spoken dialogue system, experimenting reinforcement learning based dialog management methods in real-world health applications, demonstrating feasibility of modeling relatively long (in terms of the number of dialogue turns and the number of inputs required to receive) dialogue interactions with Markov Decision Processes (MDPs) and partially observable MDPs.