Date of Award

2016

Degree Type

Thesis

Degree Name

M.Sc.

Department

Computer Science

First Advisor

YUAN, XIAOBU

Keywords

COCOM Model, Embodied conversational agents, Facial expression modeling, Fuzzy Logic, Human computer interaction, POMDP

Rights

CC BY-NC-ND 4.0

Abstract

An Embodied Conversational Agent (ECA) is an intelligent agent that interacts with users through verbal and nonverbal expressions. When used as the front-end of software applications, these agents provide online assistance that transcends the limits of time, location, and even language. To help improve user experience, there is an increasing need to make ECA more realistic, which can be achieved by using more naturalistic facial expressions. The proposed thesis research is going to work on the modeling of facial expressions based on the changing trend in the history of belief states and fuzzy logic. This work enhances the software customization system developed previously in our research group with a modification that links the changing value of belief states with the change of emotion expressions at both the mode and scale levels. In addition, the operation of the proposed method is implemented in an online bookstore system, and its success is verified with a usability study.

Share

COinS