Fuzzy rule-based emotional advisor for assisting children with autism in a social communication

• Main Author: Lim, Shimin.
• Other Authors: Cho Siu-Yeung David
• Format: Final Year Project
• Language: English
• Published: 2011
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer applications::Social and behavioral sciences
• Online Access: http://hdl.handle.net/10356/44145
• Institution: Nanyang Technological University

Children diagnosed with autism are thought to lack or possess impairment in certain representational sets of abilities. As a result, they face difficulties operating in our highly complex social environment. Specifically, they are unable to understand other people’s emotions and react accordingly. To allow children with autism to be able to react and respond in ways that other individuals are able to understand them, we propose a portable computer-based device called Fuzzy Rule-Based Emotional Advisor to assist autistic children in engaging in meaningful conversations where people are able to ascertain how they are feeling during communication. In this project, a device will be developed with camera connecting to a mobile PC. The software running on the mobile PC receives the image from camera and recognizes the emotional states and based on the emotion detected, the system generates an advice to the autistic children that suggests an appropriate reaction that they can give during communication. The system is real-time so that computation time is vital to enable real-time interactive training for the system to learn and analyze the human emotions for autistic children. This research is believed to be the first attempts based on the combination of fuzzy logic, machine learning and pattern recognition together with a neuroscience understanding of cognitive and visual signal interplay in bridging the communication chasm between autistic children and the world.