Generative Modeling of Itemset Sequences Derived from Real Databases

Rui Henriques, Cláudia Antunes

2014

Abstract

The increasingly studied problem of discovering temporal and attribute dependencies from multi-sets of events derived from real-world databases can be mapped as a sequential pattern mining task over itemset sequences. Still, the length and local nature of pattern-based models have been limiting its application. Although generative approaches can offer a critical compact and probabilistic view of sequential patterns, existing contributions are only prepared to deal with sequences of single elements. This work targets the task of modeling itemset sequences under a Markov assumption using models centered on sequential patterns. Experimental results hold evidence for the ability to model sequential patterns with acceptable completeness and precision levels, and with superior efficiency for dense or large datasets. We show that the proposed learning setting allows: i) compact representations; ii) the probabilistic decoding of patterns; iii) the inclusion of user-driven constraints through simple parameterizations; and iv) the use of the generative pattern-centered models to support key tasks such as classification. Relevance is demonstrated on retail and administrative databases.

Paper Citation

in Harvard Style

Henriques R. and Antunes C. (2014). Generative Modeling of Itemset Sequences Derived from Real Databases . In Proceedings of the 16th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-027-7, pages 264-272. DOI: 10.5220/0004898302640272

in Bibtex Style

@conference{iceis14,
author={Rui Henriques and Cláudia Antunes},
title={Generative Modeling of Itemset Sequences Derived from Real Databases},
booktitle={Proceedings of the 16th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2014},
pages={264-272},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004898302640272},
isbn={978-989-758-027-7},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 16th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - Generative Modeling of Itemset Sequences Derived from Real Databases
SN - 978-989-758-027-7
AU - Henriques R.
AU - Antunes C.
PY - 2014
SP - 264
EP - 272
DO - 10.5220/0004898302640272