Analysis of Semestral Progress in Higher Technical Education with HMM Models.

Supporting educational processes with Hidden Markov Models (HMMs) has great potential. In this paper, we explore the possibility of identifying students’ learning progress with HMMs. Students’ grades are used to train the HMMs to find out if the analysis of obtained models lets us detect patterns emerging from student’s results. We also try to predict the final students’ results on the basis of their partial grades. A new, classification approach for this problem, using properties of HMMs is proposed: High and Low State Model (HLSM).