Bijective soft set classification of student´s learning styles

Adaptive learning systems require student model repository that store student´s data in order to provide learning materials tailored to the student´s characteristics. Student´s learning style is one of the most crucial factors to be identified in order to update the student´s profile. Hence, the need for a high accurate classifier that deals with the students data and generates suitable number of rules that can automatically identify student´s learning style in adaptive learning environments is an important issue. In this paper, Bijective soft set algorithm is used in classifying students learning style based on Felder Silverman learning style model. The discernible matrix parameter reduction was used to generate attributes reducts, whereby rules were generated from these reducts. Experiments were carried out using simulated data with various attributes that represent student´s characteristics according to the four dimensions of Felder Silverman model. The experiments revealed that Bijective soft set gives high classification accuracy with suitable number of rules. The algorithm was further verified with Rough Set and FURIA as a comparative analysis of the performance of the proposed model.