论文信息 - Extraction de chroniques discriminantes

Extraction de chroniques discriminantes

L'extraction de motifs sequentiels vise a extraire des comportements recurrents dans un ensemble de sequences. Lorsque ces sequences sont etiquetees, l'extraction de motifs discriminants engendre des motifs caracteristiques de chaque classe de sequences. Cet article s'interesse a l'extraction des chroniques discriminantes ou une chronique est un type de motif temporel representant des durees inter-evenements quantitatives. L'article presente l'algorithme DCM dont l'originalite reside dans l'utilisation de methodes d'apprentissage automatique pour extraire les intervalles temporels. Les performances computationnelles et le pouvoir discriminant des chroniques extraites sont evalues sur des donnees synthetiques et reelles.

[1] Marie-Odile Cordier,et al. Temporal abstraction and inductive logic programming for arrhythmia recognition from electrocardiograms , 2003, Artif. Intell. Medicine.

[2] A. Akhmetova. Discovery of Frequent Episodes in Event Sequences , 2006 .

[3] Alain Mille,et al. A complete chronicle discovery approach: application to activity analysis , 2012, Expert Syst. J. Knowl. Eng..

[4] John F. Roddick,et al. Sequential pattern mining -- approaches and algorithms , 2013, CSUR.

[5] Christophe Dousson,et al. Discovering Chronicles with Numerical Time Constraints from Alarm Logs for Monitoring Dynamic Systems , 1999, IJCAI.

[6] Gary Doran,et al. A theoretical and empirical analysis of support vector machine methods for multiple-instance classification , 2014, Machine Learning.

[7] Johannes Fürnkranz,et al. From Local Patterns to Global Models: The LeGo Approach to Data Mining , 2008 .

[8] Thomas Guyet,et al. Extracting Temporal Patterns from Interval-Based Sequences , 2011, IJCAI.

[9] Huilong Duan,et al. On mining clinical pathway patterns from medical behaviors , 2012, Artif. Intell. Medicine.

[10] William W. Cohen. Fast Effective Rule Induction , 1995, ICML.