# Motifs séquentiels δ-libres

1 DM2L - Data Mining and Machine Learning
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
2 ORPAILLEUR - Knowledge representation, reasonning
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
3 Equipe CODAG - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : Sequential pattern mining is a challenging task with important locks like the size of the output. In this paper, we propose a new approach that extract the more general patterns and suppress the more specific patterns with similar frequencies. We defined $\delta$-sequential patterns that enable to reduce the output. Even if this notion is already known for itemsets, we show that its extension to the sequence framework is very difficult. The approach produces few and useful patterns for data mining tasks like sequence classification.
Mots-clés :
Document type :
Conference papers
Domain :

https://hal.inria.fr/hal-00653579
Contributor : Chedy Raïssi <>
Submitted on : Monday, December 19, 2011 - 5:15:59 PM
Last modification on : Friday, December 11, 2020 - 12:36:02 PM

### Identifiers

• HAL Id : hal-00653579, version 1

### Citation

Marc Plantevit, Chedy Raïssi, Bruno Crémilleux. Motifs séquentiels δ-libres. Extraction et gestion des connaissances (EGC'2011), Ali Khenchaf, Pascal Poncelet, Jan 2011, Brest, France. ⟨hal-00653579⟩

Record views