An OER Recommender System Supporting Accessibility Requirements

Author: M. Elias, M. R. Tavakoli, S. Lohmann, G. Kismihók, S. Auer
Journal: ASSETS '20: The 22nd International ACM SIGACCESS Conference on Computers and Accessibility
Year: 2020

Citation information

M. Elias, M. R. Tavakoli, S. Lohmann, G. Kismihók, S. Auer,
ASSETS '20: The 22nd International ACM SIGACCESS Conference on Computers and Accessibility,
2020,
57,
1-4,
October,
https://dl.acm.org/doi/10.1145/3373625.3418021

Open Educational Resources are becoming a significant source of learning that are widely used for various educational purposes and levels. Learners have diverse backgrounds and needs, especially when it comes to learners with accessibility requirements. Persons with disabilities have significantly lower employment rates partly due to the lack of access to education and vocational rehabilitation and training. It is not surprising therefore, that providing high quality OERs that facilitate the self-development towards specific jobs and skills on the labor market in the light of special preferences of learners with disabilities is difficult. In this paper, we introduce a personalized OER recommeder system that considers skills, occupations, and accessibility properties of learners to retrieve the most adequate and high-quality OERs. This is done by: 1) describing the profile of learners with disabilities, 2) collecting and analysing more than 1,500 OERs, 3) filtering OERs based on their accessibility features and predicted quality, and 4) providing personalised OER recommendations for learners according to their accessibility needs. As a result, the OERs retrieved by our method proved to satisfy more accessibility checks than other OERs. Moreover, we evaluated our results with five experts in educating people with visual and cognitive impairments. The evaluation showed that our recommendations are potentially helpful for learners with accessibility needs.