Article: An efficient approach for significant time intervals of frequent itemsets Journal: International Journal of Intelligent Systems Technologies and Applications (IJISTA) 2014 Vol.13 No.3 pp.222 - 243 Abstract: Ignoring the time stamps of the transactions may lead to incomplete analysis of patterns and researchers have considered the time stamps of the transactions to get more insight into frequent patterns. It is also noticed that patterns become frequent though their frequency do not spread uniformly over the entire transactional database, rather present in some part of the database might be enough. This observation motivated researchers to design algorithms to perform micro analysis over the database for better understanding of the hidden knowledge. This paper extended the frequent pattern mining framework to extract and associate certain time intervals to each of frequent itemset. We introduced significant time intervals and maximal significant time intervals and proposed an algorithm named as OTIS. Significant time interval (a, b) of a frequent itemset X is extracted based on the support of X in a subset of transactions corresponding to the time interval (a, b). Extraction of significant time intervals of frequent itemsets must be of great interest as they influence business decisions. Inderscience Publishers - linking academia, business and industry through research

Title: An efficient approach for significant time intervals of frequent itemsets

Authors: Somaraju Suvvari; R.B.V. Subramanyam

Addresses: Department of Computer Science and Engineering, National Institute of Technology, Warangal, Warangal 506004, India ' Department of Computer Science and Engineering, National Institute of Technology, Warangal, Warangal 506004, India

Abstract: Ignoring the time stamps of the transactions may lead to incomplete analysis of patterns and researchers have considered the time stamps of the transactions to get more insight into frequent patterns. It is also noticed that patterns become frequent though their frequency do not spread uniformly over the entire transactional database, rather present in some part of the database might be enough. This observation motivated researchers to design algorithms to perform micro analysis over the database for better understanding of the hidden knowledge. This paper extended the frequent pattern mining framework to extract and associate certain time intervals to each of frequent itemset. We introduced significant time intervals and maximal significant time intervals and proposed an algorithm named as OTIS. Significant time interval (a, b) of a frequent itemset X is extracted based on the support of X in a subset of transactions corresponding to the time interval (a, b). Extraction of significant time intervals of frequent itemsets must be of great interest as they influence business decisions.

Keywords: FPM; frequent pattern mining; significant time intervals; insignificant time intervals; extended significant time intervals; maximal significant time intervals; ordered time interval sets; frequent itemsets; frequent patterns; data mining.

DOI: 10.1504/IJISTA.2014.065181

International Journal of Intelligent Systems Technologies and Applications, 2014 Vol.13 No.3, pp.222 - 243

Received: 04 Nov 2013
Accepted: 15 May 2014
Published online: 15 Oct 2014 *

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

Title: An efficient approach for significant time intervals of frequent itemsets

Keep up-to-date