Naslov
A Method for Automatic Selection and Interpretation of Student Clustering Models According to their Activity on e-learning System

Autori
Jugo, Igor ; Kovačić, Božidar

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Central European Conference on Information and Intelligent Systems / Strahonja, Vjeran ; Kirinić, Valentina - Varaždin : Faculty of Organisation and Informatics, Varazdin, 2017, 61-68

Skup
28th Central European Conference on Information and Intelligent Systems

Mjesto i datum
Varaždin, Hrvatska, 27.09.2017. - 29.09.2017

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Clustering, Educational data mining, Intelligent tutoring system, E-learning

Sažetak
The paper proposes a method that is part of a new, extended architecture of our web-based intelligent tutoring system (ITS). It was developed to provide hints to students during learning through the application of educational data mining (EDM). The architecture consists of three modules – a) a communication module that enables seamless communication with data mining tools ; b) a clustering module that discovers clusters in student data based on their activity and c) a sequential pattern mining (SPM) module that finds efficient frequent learning patterns of students in each cluster. Finally, the obtained results are used by the tutoring module to provide hints to a student on which item to learn next (or previous to the selected one). To improve the hint selection process we developed a method for cluster grading to determine which cluster represents the group that has been using the ITS in the manner closest to an envisioned optimum. We verify the method on data gathered from two groups of students who used the system to master a knowledge domain, and present the obtained results.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

POVEZANOST RADA