Naslov
Providing Hints Based On Discovered Frequent High- Utility Patterns In A Web-Based ITS

Autori
Jugo, Igor ; Kovačić, Božidar

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

Izvornik
Proceedings of 8th Conference on e-learning / Jovanović, Slobodan ; Trebinjac, Bojana ; Kovačević, Sanja - Beograd : Belgrade Metropolitan University, 2017, 87-92

Skup
8th Conference on eLearning

Mjesto i datum
Beograd, Srbija, 28.09.2017. - 29.09.2017

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Sequence mining, providing hints, high-utility paths, student guidance

Sažetak
This paper presents an exploratory data mining methodology for discovering frequent high- utility learning patterns and using them to provide hints in a web-based intelligent tutoring system (ITS). This is achieved through several phases, namely: obtaining and evaluating learning traces of each student ; transforming and encoding these traces in order to create a set of paths that led to a learning unit (“prefix” paths) and a set of paths a student took after learning a unit (“suffix” paths) ; combining the paths of all students for each learning unit into a final dataset, performing high-utility sequential pattern mining and finally storing the discovered frequent patterns in the systems database. This enables the final goal of the system – offering hints to students using our ITS. We present and discuss the results obtained by applying this methodology on interaction data from two different knowledge domains.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

POVEZANOST RADA