INSA Lyon contribution to cInQ IST-2000-26469

cInQ, project IST-2000-26469, is  funded partially by the Future and Emergent Technologies arm of the IST programme FET-Open scheme. This consortium is made of the Universita degli Studi di Torino (Italy, contact Rosa Meo), the Politecnico di Milano (Italy, contact Pier-Luca Lanzi and Stefano Ceri), the Albert-Ludwigs Universitaet Freiburg (Germany, contact Luc de Raedt), the Nokia Research Center in Helsinki (Finland, contact Mika Klemettinen and Heikki Mannila), the Institute Jozef Stefan (Slovenia, contact Saso Dzeroski) and INSA Lyon (France, coordinator: Jean-François Boulicaut). The project has started on May 1st, 2001 and it stops on April 30th, 2004.

During the two first years of the project, the major contribution to cInQ for the INSA partner has concerned:

• The study of query language issues
• The theoretical framework of epsilon adequate representations for frequency queries. It has lead to several condensed representations of the collection of the frequent itemsets in transactional databases
• The design of efficient solvers for itemset and sequence constraint-based mining
• The study of the multiple uses of the frequent itemsets
• Several applications of frequent patterns (and their condensed representations) for gene expression data analysis and log analysis

Participants

Academics (paid by INSA Lyon)

Jean-François Boulicaut, Associate Professor (HdR) at INSA Lyon, cInQ project coordinator
Christophe Rigotti, Associate Professor at INSA Lyon

Other contributors

We provide here the list of the researchers that have been involved in scientific tasks related to the cInQ worplan. With the exception of Cyrille Masson who is paid by the cInQ project, the other researchers have been paid on other sources.

Hunor Albert-Lorincz, Master of Science (July 2002)
Jérémy Besson, Ph.D. student since October 2002)
Sylvain Blachon, Ph.D. student since December 2002)
Artur Bykowski, Ph.D. (October 2002)
Baptiste Jeudy, Ph.D. (December 2002)
Matthieu Capelle, Master of Science (September 2001)
Bruno Crémilleux, Associate professor at the University of Caen (invited researcher at INSA Lyon from February until July 2001)
Cyrille Masson, Ph.D. student since June 2001
François Rioult, Ph.D. student (University of Caen)
Céline Robardet, Ph. D. (July 2002)

Participation to administrative coordination

Claire Leschi, Associate Professor at INSA Lyon
Sandrine Ranieff, secretary
Anne Tchounikine, Associate Professor at INSA Lyon

Survey of contribution

We report here the publications that have been associated to cInQ deliverables, i.e., for which the research has been partially funded by the FET arm of the IST programme (European Commission)

Publications

International refeered journals

 C. Becquet, S. Blachon, B. Jeudy, J-F. Boulicaut, O. Gandrillon. Strong-association-rule mining for large-scale gene-expression data analysis: a case study on human SAGE data. Genome Biology 3(12) 2002..

B. Jeudy, J-F. Boulicaut. Optimization of association rule mining queries. Intelligent Data Analysis journal, 6 (4) 2002. pp. 341-357. IOS Press.

J-F. Boulicaut, A. Bykowski, C. Rigotti. Free-sets: a condensed representation of boolean data for frequency query approximation. Data Mining and Knowledge Discovery journal 7 (1) 2003. pp 5-22. Kluwer Academics Publishers.

A . Bykowski, C. Rigotti. DBC: a condensed representation of frequent patterns for efficient mining. Information Systems Vol. 28, Number 8, December 2003. pp. 949-977. Elsevier Science.

J. Besson, C. Robardet, J-F. Boulicaut, S. Rome. Constraint-based concept mining and its application to microarray data analysis. Accepted for publication in the Intelligent Data Analysis journal in January 2004. IOS Press.

Chapters in books

M. Botta, J-F. Boulicaut, C. Masson, R. Meo. Query languages supporting descriptive rule mining: a comparative study. Database support for Data Mining Applications, R. Meo et al. Eds., Springer-Verlag LNCS 2682. pp. 27-56. This paper is a result on an intra-cInQ cooperation between the University of Torino and INSA Lyon.

J-F. Boulicaut. Inductive databases and multiple uses of frequent itemsets: the cInQ approach. Database support for Data Mining Applications, R. Meo et al. Eds., Springer-Verlag LNCS 2682, pp. 3-26.

International conferences and workshops (refeered full papers)

A. Bykowski and C. Rigotti. A condensed representation to find frequent patterns. In: Proceedings of the joint conference ACM SIGMOD-PODS 2001, May 21-24, 2001, Santa Barbara (USA). pp. 267-273.

J-F. Boulicaut, B. Jeudy. Mining Free Sets under Constraints.In: Proceedings of the International Database Engineering and Applications Symposium IDEAS'01, Grenoble (F), July 2001, IEEE Computer Press. pp. 322-329.
 
J-F. Boulicaut, B. Jeudy. Constraint-based discovery of a condensed representation for frequent patterns. In: Proceedings of the Workshop Database Support for KDD co-located with the 5th European Conference on Principles and Practice of Knowledge Discovery in Databases PKDD'01, Freiburg (D), September 7, 2001. pp. 3-13. Available on line.

 J-F. Boulicaut, B. Crémilleux. Delta-strong classification rules for predicting collagen diseases. In: Proceedings of the PKDD'01 Discovery Challenge on Thrombosis Data co-located with the 5th European Conference on Principles and Practice of Knowledge Discovery in Databases PKDD'01, Freiburg (D), September 6, 2001. pp. 29-38. Available on line.

J-F. Boulicaut, B. Crémilleux. Delta-strong classification rules for characterizing chemical carcinogens. In: Proceedings of the Predictive Toxicology Challenge 2000-2001 co-located with the 5th European Conference on Principles and Practice of Knowledge Discovery in Databases PKDD'01, Freiburg (D), September 6, 2001. 8 p. Available on line.

M. Botta, J-F. Boulicaut, C. Masson, R. Meo. A comparison between query languages for the extraction of association rules. in: Proceedings of the Fourth International Conference on Data Warehousing and Knowledge Discovery DaWaK'02, Aix-en-Provence (F), September 4-6, 2002. Springer-Verlag LNCS volume 2454. pp. 1-10. This paper is a result on an intra-cInQ cooperation between the University of Torino and INSA Lyon.

C. Masson, F. Jacquenet. Mining frequent logical sequences with SPIRIT-LoG. Proceedings of  the 12th International Conference on Inductive Logic Programming ILP'02, Sydney (Australia), July 2002. Springer-Verlag LNAI volume 2583. pp. 166-182.

C. Robardet, B. Crémilleux, J-F. Boulicaut. Characterization of unsupervized clusters by means of the simplest association rules: an application for child's meningitis. in: Proceedings of the 7th Workshop on Intelligent Data Analysis in Medicine and Pharmacology IDAMAP'02, co-located with ECAI'02, Lyon (F), July 23, 2002.

M. Capelle, C. Masson, J-F. Boulicaut. Mining frequent sequential patterns under a similarity constraint. in: Proceedings of the Third International Conference on Intelligent Data Engineering and Automated Learning IDEAL 2002, Manchester (UK), August 12-14, 2002. Springer-Verlag LNCS volume 2412. pp. 1-6.

B. Jeudy, J-F. Boulicaut. Using condensed representations for interactive association rule mining. in: Proceedings of the 6th European conferences on Principles and practice of Knowledge Discovery in Databases PKDD 2002, Helsinki (FIN), 19-23 August 2002. Springer-Verlag LNAI volume 2431, pp. 225-236.

B. Jeudy, J-F. Boulicaut. Constraint-based discovery and inductive queries: application to association rule mining. in: Proceedings of the European Science Foundation Exploratory Workshop on Pattern Detection and Discovery in Data Mining, London (UK). 16-18 September 2002. Springer-Verlag LNAI volume 2447. pp. 120-124.

B. Crémilleux, J-F. Boulicaut. Simplest rules characterizing classes generated by delta-free sets. In: Proceedings of the 22nd SGAI International Conference on Knowledge Based Systems and Applied Artificial Intelligence ES 2002, Cambridge (UK), 10-12 December 2002. Springer-Verlag. pp. 33-46.

H. Albert-Lorincz, J-F. Boulicaut. Mining frequent sequential patterns under regular expressions: a highly adaptive strategy for pushing constraints. Proceedings of the 3rd SIAM International Conference on Data Mining SDM'03, San Francisco (USA), May 1-3, 2003. pp. 316-320 (see also on line proceedings).

K. Hatonen, J-F. Boulicaut, M. Klemettinen, M. Miettinen, C. Masson. Comprehensive log compression with frequent patterns. Proceedings of the 5th International Conference on Data Warehousing and Knowledge Discovery DaWaK 2003, Prague (CZ),  September 3-5, 2003. Springer-Verlag LNCS 2737. pp. 360-370. This paper is a result on an intra-cInQ cooperation between the NOKIA Research Center in Helsinki and INSA Lyon.

F. Rioult, J-F. Boulicaut, B. Crémilleux, J. Besson. Using transposition for pattern discovery from microarray data. Proceedings of the 8th ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery, June 13th, 2003, San Diego (USA), M. J. Zaki and C.C. Aggarwal Eds., June 13th, 2003, San Diego (USA). pp. 73-79.

M. Leleu, C. Rigotti, J-F. Boulicaut, G. Euvrard. Constraint-based sequential pattern mining over datasets with consecutive repetitions. Proceedings of the 7th European Conference on Principles and Practice of Knowledge Discovery in Databases PKDD 2003, Cavtat-Dubrovnik (Croatia), September 22-26, 2003  Springer-Verlag LNAI 2838. pp. 303-314.

H. Albert-Lorincz, J-F. Boulicaut. A framework for frequent sequence mining under generalized regular expression constraints. Proceedings of the 2nd International Workshop on Knowledge Discovery in Inductive Databases KDID'03 co-located with ECML-PKDD 2003, Cavtat-Dubrovnik (Croatia), September 22, 2003. pp. 2-16. On line from WWW site.

F. Rioult, C. Robardet, S. Blachon, B. Crémilleux, O. Gandrillon, J-F. Boulicaut. Mining concepts from large SAGE gene expression matrices. Proceedings of the 2nd International Workshop on Knowledge Discovery in Inductive Databases KDID'03 co-located with ECML-PKDD 2003, Cavtat-Dubrovnik (Croatia), September 22, 2003  pp. 107-118. On line from WWW site.

C. Masson, C. Robardet, J-F. Boulicaut. Optimizing subset queries: a step towards SQL-based inductive databases for itemsets. Proceedings of 2004 ACM Symposium of Applied Computing (SAC'2004), Special Track on Data Mining (DM) March 14 - 17, 2004, Nicosia, Cyprus. ACM Press. pp. 535-539.

C. Robardet, R. Pensa, J. Besson, J-F. Boulicaut. Using classification and visualization on pattern databases for gene expression data analysis. Proceedings of the International Workshop on Pattern Representation and Management PaRMa'04 co-located with EDBT 2004, Heraclion - Crete, Greece, March 18, 2004. pp. 107-118.

J. Besson, C. Robardet, J-F. Boulicaut. Constraint-based mining of formal concepts in transactional data. In: Proceedings of the 8th Pacific-Asia Conference on Knowledge Discovery and Data Mining PaKDD’04, Sydney (Australia), May 26-28, 2004. Springer-Verlag LNCS 3056, pp. 615-624.

National conferences (refeered full papers that might be available only in French)

 J-F. Boulicaut, P. Marcel, C. Rigotti. Query driven knowledge discovery via OLAP manipulations. In: Actes des 17ème Journées Bases de Données Avancées BDA'01, Agadir (Maroc), Octobre 2001. Cepadues Editions. pp. 311-323.

B. Crémilleux, J-F. Boulicaut. Utilisation de règles delta-fortes pour caractériser des classes. In: Actes du 13e Congrès Francophone AFRIF-AFIA de Reconnaissance des Formes et Intelligence Artificielle RFIA'02, Angers (F), 8-10 Janvier 2002. pp. 685-694.

M. Capelle, J-F. Boulicaut, C. Masson. Extraction de motifs séquentiels sous contrainte de similarité. In: Actes des Journées francophone d'Extraction et de Gestion de Connaissances EGC'02. Montpellier (F), janvier 2002. Hermes. pp. 65-76.

S. Blachon, C. Robardet, J-F. Boulicaut, O. Gandrillon. Extraction de régularités dans des données d'expression SAGE humaines. Actes de la journée Informatique pour l'analyse du transcriptome JPGD'03, Lyon (F), 14 Mai 2003. 18 pages.

J-F. Boulicaut, F. Rioult, J. Besson, B. Crémilleux, S. Rome. Faisabilité des extractions d'ensembles fréquents dans des données biopuces : éléments de solution. Actes de la journée Informatique pour l'analyse du transcriptome JPGD'03, Lyon (F), 14 Mai 2003. 13 pages.

Academic achievements

M. Capelle. Extraction de motifs séquentiels sous contraintes. Mémoire DEA ECD (Master thesis), Septembre 2001, 35 p.

J-F. Boulicaut. Extraction de connaissances dans les données - des méthodes ad-hoc au cadre des bases de données inductives. Mémoire d'Habilitation à Diriger des Recherches (Habilitation thesis), INSA de Lyon et Université Claude Bernard Lyon 1, Novembre 2001. 107 pages + annexes.

H. Albert-Lorincz. Extraction à granularité variable de motifs séquentiels sous contraintes. Mémoire DEA ECD (Master thesis), juin 2002, 35 p.

 A. Bykowski. Condensed representations of frequent sets: application to descriptive pattern discovery. Thèse de doctorat (Ph.D. thesis) de l'INSA Lyon présentée le 21 Octobre 2002. 186 p.

B. Jeudy. Optimisation de requêtes inductives: application à l'extraction sous contraintes de règles d'association. Thèse de doctorat (Ph.D. thesis) de l'INSA Lyon. présentée le 13 décembre 2002. 130 pages.

Other dissemination actions

Invited talks

J-F. Boulicaut. The Inductive Database framework - a long-term perspective on query languages for data mining (invited talk). In: Proceedings of the Workshop Database Support for KDD co-located with the 5th European Conference on Principles and Practice of Knowledge Discovery in Databases PKDD'01, Freiburg (D), September 7, 2001. pp. 1. Available on line.

J-F. Boulicaut. The Inductive Database framework - a long-term perspective on query languages for data mining (invited talk). In: Proceedings of the Workshop International workshop DTDM'02 co-located with EDBT'02 (slides available here).

Tutorials

• Tutorial Inductive Databases and Constraint-Based Mining (with Luc De Raedt) at ECML/PKDD 2002, August 2002. A shorter version of this tutorial has been given in French at EGC 2003.
• Tutorial Constraint-based mining: a major step towards inductive databases at IDA 2003, August 2003.

Other actions

Panelist for the Innovative Projects for Intelligent Systems in the New Century panel at ISMIS 2002, June 2002

Presentation of the cInQ project at the KD-net meeting co-located with ECML/PKDD 2002 (August 20, 2002) and the KD-net exhibition co-located with ECML-PKDD 2003 (September 23, 2003).

Participations to Program Commitees

• J-F. Boulicaut serves as one of the vice-chair of ICDM'04 PC, 4th IEEE International Conference on Data Mining, Brighton (UK), 1-4 November 2004.

• J-F. Boulicaut serves as a co-chair ECML/PKDD 2004 (with F. Esposito, F. Giannotti, and D. Pedreschi), i.e., the 15th European Conference on Machine Learning and the 8th European conferences on Principles and practice of Knowledge Discovery in Databases ECML/PKDD 2004, Pisa (I), 20-24 September 2004.

• PaRMa'04 From Data to Patterns, International Workshop on Pattern Representation and Management co-located with EDBT 2004, Heraclion - Crete, Greece, March 18, 2004 (J-F. Boulicaut)

• 2004 ACM Symposium of Applied Computing (SAC'2004), Special Track on Data Mining, Nicosia (Cyprus), 14-17 March, 2004 (J-F. Boulicaut, C. Rigotti).

• Workshop on Multi Relational Data Mining MRDM 2003 co-located with ACM SIGKDD 2003, and MRDM 2002 co-located with ACM SIGKDD 2002  (J-F. Boulicaut)

• International Workshop on Knowledge Discovery in Inductive Databases KDID'03 co-located with ECML-PKDD 2003, and KDID'02 co-located with ECML/PKDD 2002 ( (J-F. Boulicaut and C. Rigotti)
• European Conference on Principles and Practice of Knowledge Discovery in Databases PKDD 2003 , ECML/PKDD 2002 (workshop chair), and PKDD 2001  (J-F. Boulicaut). In 2004, C. Rigotti serves as a PC member for PKDD 2004.
• International Conference on Inductive Logic Programming ILP 2003, ILP 2002, and ILP'01  (J-F. Boulicaut)
• European Conference on Artificial Intelligence ECAI'02 (tutorial chair), July 2002  (J-F. Boulicaut)
• EDBT 2002 Workshop on Database Technologies for Data Mining DTDM'02, March 2002  (J-F. Boulicaut and C. Rigotti)
• ECAI 2002 Workshop on Advances in Formal Concept Analysis for Knowledge Discovery in Databases FCAKDD'02, (J-F. Boulicaut)

To contact us

INSA de Lyon
LIRIS CNRS FRE 2672 (ex LISI)
Batiment Blaise Pascal
F-69621 Villeurbanne cedex
Tel. +33 (0) 4.72.43.89.05
Fax. +33 (0) 4.72.43.87.13
Jean-Francois.Boulicaut at insa-lyon dot fr