DocumentCode :
2040849
Title :
Mining and post-mining of time stamped association rules
Author :
Mathu, T. ; Narmadha, D. ; Geetha, S.
Author_Institution :
Dept. of Comput. Sci. & Eng., Karunya Univ., Coimbatore, India
Volume :
4
fYear :
2011
fDate :
8-10 April 2011
Firstpage :
149
Lastpage :
153
Abstract :
We study the problem of discovering time stamped association rules and to post-mine it for reducing the number of rules to make it user interesting. Most works in mining association rules produces a large number of rules which are almost non-interesting to the user. In this paper, we propose an algorithm which uses the concept of lower bounding distance and ontology to mine and post-mine the rules respectively. This paper defines a matching operator for rule selection. The matching operator selects the time stamped association rule that matches with the user-specified constraint. The experimental results show that the execution time is reduced when compared with the traditional mining method. Also the number of rules is greatly reduced when the matching operator is used. Our approach will help the decision maker to analyze the results effectively.
Keywords :
data mining; database management systems; ontologies (artificial intelligence); transaction processing; matching operator; ontology; post mining; rule selection; time stamped association rule mining; transactional database; Algorithm design and analysis; Association rules; Itemsets; Ontologies; Taxonomy; Time stamped association rule; matching operator; ontology; user-constraint template;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronics Computer Technology (ICECT), 2011 3rd International Conference on
Conference_Location :
Kanyakumari
Print_ISBN :
978-1-4244-8678-6
Electronic_ISBN :
978-1-4244-8679-3
Type :
conf
DOI :
10.1109/ICECTECH.2011.5941876
Filename :
5941876
Link To Document :
بازگشت