In the context of multiple intelligences theory, intelligent data analysis of learning styles was based on rough set theory

The present study aims to identify the relationship between individuals' multiple intelligence areas and their learning styles with mathematical clarity using the concept of rough sets which is used in areas such as artificial intelligence, data reduction, discovery of dependencies, prediction of data significance, and generating decision (control) algorithms based on data sets. Therefore, first multiple intelligence areas and learning styles of 243 mathematics prospective teachers studying at a state university were identified using the “Multiple Intelligence Inventory for Educators” developed by Armstrong and the “Learning Styles Scale” developed by Kolb. Second, the data was appropriated for rough set analysis and we identified potential learning styles that a student can have based on the learning style s/he already has. Certainty degrees of the learning style sets were αR(D) ≅ 0.717, αR(C) ≅ 0.618, αR(AS) ≅ 0.699, αR(AC) ≅ 0.461, and these sets were found to be rough sets. Finally, decision rules were identified for multiple intelligences and learning styles.

► We identified potential learning styles that a student can have. ► Certainty degrees of the learning style sets were determined. ► All multiple intelligences explained learning styles with a dependency degree of .794. ► Data could not be removed from the attributes set of intelligence areas. ► The decision rules were identified for multiple intelligences and learning styles.