This website presents our work on the use of ASP for pattern mining. You can find our articles and download datasets that have been used to benchmark our approach.

This work is a join work mainly between:


Our objective is to resolve a sequential pattern mining task. This task consists in extracting all the subsequences that frequently occur in a dataset of sequences. These sequences are very common. They can be functioning traces of a electronic devise, a weblog, the customer purchases or the patient care pathways. The sequential pattern mining task aims at identifying recurrent behaviors in such kind of data.

Many algorithms have been proposed to address this problem, and most of them yields a huge number of patterns in a short time. The problem is that with such a large amount of results, he can not more get insightful information than if he directly explores the data. The solution relies in the definition of additional constraints on the patterns expected by the data analyst.

Our approach aims at using the ASP framework to design a flexible pattern mining approach, ie an approach for which it will be easy to add constraints on the patterns.

The "encoding and solving example" page gives an example of a simple ASP encoding of the sequential pattern mining task and explain how to use it with the clingo solver. The other pages present the results and the datasets we recently published.


    1. Philippe Besnard and Thomas Guyet: Declarative mining of negative sequential patterns. Declarative Problem Solving Workshop DPSW@ECAI (2020),

    2. Guyet, T., Moinard, Y., and Quiniou, R., and Schaub, T. (2017) "Efficiency Analysis of ASP Encodings for Sequential Pattern Mining Tasks". Advances in Knowledge Discovery and Management, Vol. 7 : p. 41-81.

    3. Samet, A., Guyet T. and Negrevergne B. (2017), "Mining rare sequential patterns with ASP". Late breaking papers of Inductive Logic Programming conference : 51-60. (see dedicated page)

    4. Guyet T., Dauxais Y. and Happe A. (2017). "Declarative sequential pattern mining of care pathways". Proceedings of Conference on Artificial Intelligence Medecine in Europe: p. 261-266.

    5. Gebser, M., Guyet, T., Quiniou, R., Romero, J., and Schaub, T. (2016). Knowledge-based sequence mining with asp. In Proceedings of International Join Conference on Artificial Intelligence (see dedicated page)

    6. Guyet, T., Moinard, Y., Quiniou, R. and Schaub T. (submitted for LNAI special issue of EGC conference) Efficiency Analysis of ASP Encodings for Sequential Pattern Mining Tasks (see dedicated page)

    7. Guyet, T., Moinard, Y., and Quiniou, R. (2014). Using answer set programming for pattern mining. In Proceedings of conference “Intelligence Artificielle Fondamentale” (IAF). http://arxiv.org/pdf/1409.7777


    • Inria-IRISA LACODAM Team: T. Guyet, R. Quiniou, Y. Moinard, A. Samet and B. Negrevergne

    • University Potsdam (Potassco Team) : T. Schaub, M. Gebser, J. Romero