A parameter-free approach for mining robust sequential classification rules - Université de La Réunion Access content directly
Conference Papers Year : 2015

A parameter-free approach for mining robust sequential classification rules

Elias Egho
  • Function : Author
  • PersonId : 768153
  • IdRef : 180812815
Dominique Gay
Marc Boullé
  • Function : Author
  • PersonId : 906624
  • IdRef : 126326223


—Sequential data is generated in many domains of science and technology. Although many studies have been carried out for sequence classification in the past decade, the problem is still a challenge; particularly for pattern-based methods. We identify two important issues related to pattern-based sequence classification which motivate the present work: the curse of parameter tuning and the instability of common interestingness measures. To alleviate these issues, we suggest a new approach and framework for mining sequential rule patterns for classification purpose. We introduce a space of rule pattern models and a prior distribution defined on this model space. From this model space, we define a Bayesian criterion for evaluating the interest of sequential patterns. We also develop a parameter-free algorithm to efficiently mine sequential patterns from the model space. Extensive experiments show that (i) the new criterion identifies interesting and robust patterns, (ii) the direct use of the mined rules as new features in a classification process demonstrates higher inductive performance than the state-of-the-art sequential pattern based classifiers.
Fichier principal
Vignette du fichier
egho_ICDM15.pdf (688.32 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-01395002 , version 1 (10-11-2016)



Elias Egho, Dominique Gay, Marc Boullé, Nicolas Voisine, Fabrice Clérot. A parameter-free approach for mining robust sequential classification rules. IEEE International Conference on Data Mining, 2015, Atlantic City, United States. ⟨10.1109/ICDM.2015.87⟩. ⟨hal-01395002⟩
154 View
315 Download



Gmail Mastodon Facebook X LinkedIn More