Zailani, Abdullah and Tutut, Herawan and Noraziah, Ahmad and Rozaida, Ghazali and Mustafa, Mat Deris (2014) Mining Indirect Least Association Rule from Students’ Examination Datasets. Computational Science and Its Applications, 8584. pp. 783-797. ISSN 978-3-319-09152-5 (ISBN). (Published)
Full text not available from this repository. (Request a copy)Abstract
Association rule mining (ARM) is one of the most important and well researched area in data mining. Indirect association rule, a part of ARM, provides a different perspective in identifying the most useful infrequent patterns. Specifically, it refers to the property of high dependencies between two items that are rarely appeared together but indirectly occurred through another items. Besides generating nontrivial information, it also can implicitly reveal a new fact of relationship which cannot be directly determined using the typical interestingness measures. Therefore, in this paper we applied our novel algorithm called Mining Lease Association Rule (MILAR) and our measure called Critical Relative Support (CRS) to mine the indirect least association rule from the students’ examination datasets. The experimental results show that the numbers of extracted indirect association rules are reduced when the threshold value of CRS is increased. This number is also lesser than the least association rule. In addition of decreasing the number of uninteresting rules, the obtained information also can be used by educators as a basis to improve their teaching and learning strategies in the future.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | Data mining; Indirect; Least association rule; Algorithm |
Subjects: | Q Science > QA Mathematics > QA76 Computer software |
Faculty/Division: | Faculty of Computer System And Software Engineering |
Depositing User: | Mrs. Neng Sury Sulaiman |
Date Deposited: | 12 Aug 2016 06:13 |
Last Modified: | 02 Feb 2018 07:28 |
URI: | http://umpir.ump.edu.my/id/eprint/8794 |
Download Statistic: | View Download Statistics |
Actions (login required)
View Item |