Title: An improved K-prototype-based cluster algorithm for mixed educational data mining

Authors: Yuan Wang; Liping Yang; Jun Wu; Shaomiao Chen

Addresses: College of Humanities and Law, Beijing University of Chemical Technology, Beijing 100029, China; School of Economics and Management, Beijing University of Chemical Technology, Beijing 100029, China ' School of Economics and Management, Beijing University of Chemical Technology, Beijing 100029, China ' School of Economics and Management, Beijing University of Chemical Technology, Beijing 100029, China ' School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan, 411201, China

Abstract: Educational data mining is one of the important ways to improve the quality of education. Clustering is a classic data mining method. It is often applied to single-attribute data. Educational data which contains multiple types of attributes. This brings great challenges to the dissimilarity measurement of clustering algorithms. In this paper, we first collect a large amount of student education data with 18 different types of attributes. Then, based on the K-prototype algorithm framework, we improve the dissimilarity measurement method so that it can be applied to the mixed attribute characteristics of education data. Finally, a Kruskal-Wallis test is used to determine the significance of clustering properties. Experimental results show that the proposed method can obtain higher quality clustering than the state-of-the-art clustering methods.

Keywords: clustering algorithm; educational data mining; EDM; mixed attribute; learning analysis; campus big data.

DOI: 10.1504/IJES.2022.127166

International Journal of Embedded Systems, 2022 Vol.15 No.5, pp.448 - 456

Received: 16 Jun 2022
Received in revised form: 16 Jul 2022
Accepted: 18 Jul 2022
Published online: 23 Nov 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article