Title: Machine Learning to Augment Capabilities of Autistic Learners: Draw Me a Sentence
Stream: Innovation & Technology
Presentation Type: Oral Presentation
Authors:
Kamaran Fathulla, University of Lincoln, United Kingdom
Jamie Smith, University of Lincoln, United Kingdom
Abstract:
Autistic learners face considerable pressures during their education years impacting their overall well-being both physically and psychologically. One of the key challenges they face is the speed and quality of their writing skills, and the research proposed here advocates an innovative new way of helping those learners. The premise of the approach is to identify, capture skills in areas of strengths such as drawing and then with the aid of machine learning and character recognition techniques map those skills to other areas of the condition which are difficult or problematic such as writing. The proposed machine learning algorithm would map drawing actions into corresponding alphabetic representations. It is expected that learners would be encouraged to practice their writing more when they see how fluid their actions can be translated into words and sentences. The more they practice the more the algorithm will learn and the better its responses will be. The profound implications of this approach in terms of learning, attitudes, and psychological impact are considered. This approach is motivated by works such as by Mottron (2017) and Greenspan (2006).
Conference Comments & Feedback
Place a comment using your LinkedIn profile[wplinkpress_comments]
Share this Presentation
