Using Machine Learning Advances to Unravel Patterns in Subject Areas and Performances of University Students with Special Educational Needs and Disabilities (MALSEND): A Conceptual Approach

Abstract

Universities and colleges in the UK welcome about 30,000 students with special needs each year. Research shows that the dropout rate for disabled students is much higher at 31.5% when compared with about 12.3% for non-disabled students in the EU. Supporting young students with special educational needs while pursuing higher education is an ambitious and important role, which needs to be adopted by tertiary education providers worldwide. We propose, MALSEND, a conceptual platform based on human-machine intelligence (HMI), a collective intelligence of human and machine to understand patterns of learning of disabled students in higher education. This platform aims to accommodate and analyse data sets features of universities activities to discover trends in performances with regards to subject areas for autistic students, dyslexic students and students having attention deficit hyperactive disorder (ADHD), among others. Analysis of variables, such as students’ performances in modules, courses and other engagement-indices will give new insights into research questions, career advice and institutional policymaking. This paper describes the developmental activities of the MALSEND concept in phases.