Research on children's cognitive development for learning disabilities using recommendation method

Abstract

AbstractIn order to achieve effective cognitive training that can significantly improve the learning abilities of children with learning disabilities, we consider the computer‐based assistive technologies and propose a personalized cognitive training items recommendation method based on children's preferences and evaluation values. First, we devise an improved cognitive ability evaluation method to evaluate children's cognitive ability values. Second, we transform these values to corresponding cognitive levels using K‐means clustering algorithm and form comprehensive cognitive level maps to reveal the strengths and weaknesses of children in terms of cognitive abilities. Finally, an improved collaborative filtering recommendation method combining children's preferences and evaluation values are represented to generate a superior training proposal. The proposed method is implemented on our cognitive development platform (CogDaily) which is an assistive system and used for testing and practicing. The fifth‐grade students with learning disabilities in mathematics in a primary school in Hangzhou are selected as subjects. They are divided into test group and control group for 2 weeks of testing through this platform. The analysis results of the changes of cognitive abilities before and after the experiment show that the accurate recommendation proposal can effectively enhance children's cognitive abilities and improve their learning abilities.