Q123 : An adaptive learning game for autistic children using reinforcement learning and fuzzy logic
Thesis > Central Library of Shahrood University > Computer Engineering > MSc > 2017
Authors:
Amir H. Khabbaz [Author], Ali Pouyan[Supervisor], Mansoor Fateh[Advisor], Vahid Abolghasemi[Advisor]
Abstarct: Autism spectrum disorder (ASD), is a set of variable features. These features are characterized by anomalies in reciprocal social interactions and perception disorders of communication patterns; and by restricted repetitive interests and activities. These abnormalities are pervasive in all situations, but may vary in degree. Computer games has positive effects on children with ASD. Serious games had been used to enhance the conversation ability of these children. The term "serious game" refers to a game which have an explicit and carefully, educational goal and are not intended to be played primarily for entertainment. In the last decade, some serious games had been proposed for educating and therapy of children with ASD. Many tools were utilized by the mentioned games, such as eye tracker devices and robots. These tools are very expansive and unaffordable for the most of the families. Some of researches used an adaptive behavior. These systems adapt themselves to the situations, skills or mental states of the patient. This behavior would help to ease the game for the patient and increase the therapy effects of the game. This thesis presents an adapted serious game for rating social ability in children with ASD. The proposed serious game uses reinforcement learning concepts for being adaptive. It is baxsed on fuzzy logic to evaluate the social ability level of the children with ASD. The game adapts itself to the level of the autistic patient by reducing or increasing the challenges in the game via an intelligent agent during the play time. If autistic patient's communication level grows during the playtime, the challenges of game may become harder to make a dynamic procedure for evaluation. At each step or state, using fuzzy logic, the level of the plaxyer is estimated baxsed on some attributes such as Health and number of the times that main character had been hit by the enemies. The proposed serious game is tested with taking part of 15 children with ASD. baxse on the obtained data, an average vector of their function is created. This vector had a positive 11 degree slope. This slope shows that the proposed method is functional and might be helpful in long term.
Keywords:
#adaptive game #reinforcement learning #fuzzy logic #ASD #serious game Link
Keeping place: Central Library of Shahrood University
Visitor: