Fast Determination of Items Support Technique from Enhanced Tree Data Structure

Abdullah, Zailani and Herawan, Tutut and Noraziah, Ahmad and Mustafa, Mat Deris (2014) Fast Determination of Items Support Technique from Enhanced Tree Data Structure. International Journal of Software Engineering and Its Applications (IJSEIA), 8 (1). pp. 21-32. ISSN 1738 - 9984. (Published)

[img] PDF
2.pdf - Published Version
Restricted to Repository staff only

Download (343kB) | Request a copy

Abstract

Frequent Pattern Tree (FP-Tree) is one of the famous data structure to keep frequent itemsets. However when the content of transactional database is modified, FP-Tree must be reconstructed again due to the changes in patterns and items support. Until this recent, most of the techniques in frequent pattern mining are using the original database to determine the items support and not from their recommended trees data structure. Therefore in this paper, we proposed a technique called Fast Determination of Item Support Technique (F-DIST) to capture the items support from our suggested Disorder Support Trie Itemset (DOSTrieIT) data structure. Experiments with the UCI datasets show that the processing time to determine the items support using F-DIST from DOSTrieIT is outperformed the classical FP-Tree technique. Furthermore, the processing time to construct a complete tree data structure for DOSTrieIT is lesser than the benchmarked CanTree data structure.

Item Type: Article
Uncontrolled Keywords: Association rules; Frequent pattern; Tree structure; Fast technique
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty/Division: Faculty of Computer System And Software Engineering
Depositing User: PM Dr. Noraziah Ahmad
Date Deposited: 22 Sep 2014 06:47
Last Modified: 01 Feb 2018 23:53
URI: http://umpir.ump.edu.my/id/eprint/6612
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item