Evaluating the application of interactive classification system in university study course comparison

Abstract

Large amount of routine work necessary to perform comparative analysis of university courses possess the importance of the automation of this task. Possibility of effortless detection of similarity between different courses would give the opportunity to organize student exchange programmes effectively and facilitate curriculum management and development. The application of smartly adapted machine learning technologies in long term could reduce the manual course comparison effort. The goal of this paper is to present the application of earlier proposed inductive learning based classification system (accompanied with interactive capabilities) to directly and indirectly compare study courses semi-automatically. The evaluation of the proposed system has been carried out in 4 consecutive experiments which proved the ability to decrease the number of misclassified instances if uncertain classifications are detected and passed to the expert's review.