Charles Darwin University

CDU eSpace
Institutional Repository

 
CDU Staff and Student only
 

A Clustering Method using Entropy for Grouping Students

Kim, Byoung Wook, Mason, Jon and Shon, Jin Gon (2015). A Clustering Method using Entropy for Grouping Students. In: Kojiri, Tomoko, Supnithi, Thepchai, Wang, Yonggu, Wu, Ying-Tien, Ogata, Hiroaki, Chen, Weiqin, Kong, Siu Cheung and Qiu, Feiyue 23rd International Conference on Computers in Education ICCE 2015, Hangzhou, China, 30 November - 4 December 2015.

Document type: Conference Paper
Citation counts: Google Scholar Search Google Scholar

IRMA ID 84550754xPUB69
Author Kim, Byoung Wook
Mason, Jon
Shon, Jin Gon
Title A Clustering Method using Entropy for Grouping Students
Conference Name 23rd International Conference on Computers in Education ICCE 2015
Conference Location Hangzhou, China
Conference Dates 30 November - 4 December 2015
Conference Publication Title Workshop Proceedings of the 23rd International Conference on Computers in Education ICCE 2015
Editor Kojiri, Tomoko
Supnithi, Thepchai
Wang, Yonggu
Wu, Ying-Tien
Ogata, Hiroaki
Chen, Weiqin
Kong, Siu Cheung
Qiu, Feiyue
Place of Publication Japan
Publisher ICCE 2014 Organizing Committee - ICT Unit, Center for Graduate Education Initiative, Japan Advanced Institute of Science and Technology
Publication Year 2015
ISBN 978-4-9908014-7-2   (check CDU catalogue open catalogue search in new window)
Start Page 418
End Page 422
Total Pages 5
HERDC Category E1 - Conference Publication (DIISR)
Abstract This study suggests a novel clustering method using entropy in information theory for setting cut-scores. Based on item response vectors from the examinees, we construct the Ordered Item Booklets (OIBs) based on the Rasch model which is a kind of Item Response Theory (IRT). The approach of the proposed method is to partition the scores into n-clusters and to construct probability distribution tables separately for each cluster from the item response vector. Using these probability distribution tables, mutual information and relative entropy (Kullback-leibler divergence) were computed between each of the clusters and then cut-scores
were determined by the cluster’s partition to minimize mutual information values. Experimental results show that the approach of this proposed entropy method has a realistic possibility of application as a clustering evaluation method
Keyword Entropy
Clustering
Item response data
Test score
Description for Link Link to conference paper
Link to conference proceedings
URL http://www.apsce.net/uploaded/filemanager/6f0bc78c-8102-4a3b-8eb3-3d01213d6d58.pdf
http://www.apsce.net/icces.php?id=f51279ea-b952-4aeb-b5bb-1b2256ff48dd
 
Versions
Version Filter Type
Access Statistics: 61 Abstract Views  -  Detailed Statistics
Created: Tue, 26 Jul 2016, 12:11:48 CST