DISCOVERING N-ARY TIMED RELATIONS FROM SEQUENCES

Nabil Benayadi, Marc Le Goc

2010

Abstract

The goal of this position paper is to show the problems with most used timed data mining techniques for discovering temporal knowledge from a set of timed messages sequences. We will present from a simple example that Apriori-like algorithms for mining sequences as Minepi and Winepi fail for mining a simple sequence generated by a very simple process. Consequently, they cannot be applied to mine sequences generated by complexes process as blast furnace process. We will show also that another technique called TOM4L(Timed Observations Mining for Learning) can be used for mining such sequences and generate significantly better results than produced by Apriori-like techniques. The results obtained with an application on very complex real world system is presented to show the operational character of the TOM4L.

Paper Citation

in Harvard Style

Benayadi N. and Le Goc M. (2010). DISCOVERING N-ARY TIMED RELATIONS FROM SEQUENCES . In Proceedings of the 2nd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-674-021-4, pages 428-433. DOI: 10.5220/0002762304280433

in Bibtex Style

@conference{icaart10,
author={Nabil Benayadi and Marc Le Goc},
title={DISCOVERING N-ARY TIMED RELATIONS FROM SEQUENCES},
booktitle={Proceedings of the 2nd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2010},
pages={428-433},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002762304280433},
isbn={978-989-674-021-4},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 2nd International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - DISCOVERING N-ARY TIMED RELATIONS FROM SEQUENCES
SN - 978-989-674-021-4
AU - Benayadi N.
AU - Le Goc M.
PY - 2010
SP - 428
EP - 433
DO - 10.5220/0002762304280433