INTERTRASM - A Depth First Search Algorithm for Mining Intertransaction Association Rules

Dan Ungureanu, Alexandru Boicea

2008

Abstract

In this paper we propose an efficient method for mining frequent intertransaction itemsets. Our approach consists in mining maximal frequent itemsets (MFI) by extending the SmartMiner algorithm for the intertransaction case. We have called the new algorithm InterTraSM (Inter Transaction Smart Miner). Because it uses depth first search the memory needed by the algorithm is reduced; a strategy for passing tail information for a node combined with a dynamic reordering heuristic lead to improved speed. Experiments comparing InterTraSM to other existing algorithms for mining frequent intertransaction itemsets have revealed a significant gain in performance. Further development ideas are also discussed.

Paper Citation

in Harvard Style

Ungureanu D. and Boicea A. (2008). INTERTRASM - A Depth First Search Algorithm for Mining Intertransaction Association Rules . In Proceedings of the Third International Conference on Software and Data Technologies - Volume 3: ICSOFT, ISBN 978-989-8111-53-1, pages 148-153. DOI: 10.5220/0001880701480153

in Bibtex Style

@conference{icsoft08,
author={Dan Ungureanu and Alexandru Boicea},
title={INTERTRASM - A Depth First Search Algorithm for Mining Intertransaction Association Rules},
booktitle={Proceedings of the Third International Conference on Software and Data Technologies - Volume 3: ICSOFT,},
year={2008},
pages={148-153},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001880701480153},
isbn={978-989-8111-53-1},
}

in EndNote Style

TY - CONF
JO - Proceedings of the Third International Conference on Software and Data Technologies - Volume 3: ICSOFT,
TI - INTERTRASM - A Depth First Search Algorithm for Mining Intertransaction Association Rules
SN - 978-989-8111-53-1
AU - Ungureanu D.
AU - Boicea A.
PY - 2008
SP - 148
EP - 153
DO - 10.5220/0001880701480153