Application of Rough Set Theory to Improve the Efficiency of Higher Education Systems

The purpose of this paper is to show the possibility of application of the Rough Set Theory to improve the efficiency of the higher education system. The study used a dataset describing with 35 attributes 4424 examples. The attributes include information known at the time of student enrollment (academic path, demographics, and social-economic factors) and the students’ academic performance at the end of the first and second semesters. The data set was explored using the Rough Set Theory to discover knowledge (recurrence and dependencies in data) represented by decision rules. The attempts to induce decision rules revealed interesting relationships between the characteristics of the examples included in the decision table and the students’ success (graduate) or failure (dropout). The relationships take a largely readable and easily interpretable form: if the premise, then the conclusion. The procedure carried out can be easily adapted to the conditions and needs of higher education systems. It will facilitate monitoring the characteristics of students, inferring in the form of clear rules and possibly taking measures directed at reducing the likelihood of students dropping out. This will make it possible to raise the effectiveness of the higher education system (fewer students dropping out) while increasing efficiency (early prediction of risks and spending on measures precisely aimed at eliminating risks). The study verifies the usefulness of the Rough Set Theory to build a rule base for decision-making in order to increase the efficiency and effectiveness of higher education.