Temporal and Frequential Metric Learning for Time Series kNN Classification

Cao-Tri Do; Ahlame Douzal-Chouakria; Sylvain Marié; Michèle Rombaut

Temporal and Frequential Metric Learning for Time Series kNN Classification

Cao-Tri Do ^{1, 2} Ahlame Douzal-Chouakria ² Sylvain Marié ³ Michèle Rombaut ¹

1 GIPSA-AGPIG - AGPIG

GIPSA-DIS - Département Images et Signal

2 LIG - Laboratoire d'Informatique de Grenoble

3 SE - Schneider Electric

Abstract : In order to classify time series, many machine learning algorithms such as the kNN classifer require a metric. We propose in this work a framework to learn a combination of multiple metrics for a robust kNN classifier. This combined metric includes both temporal (value and behavior) and frequential components. By introducing the concept of pairwise space, the combination function is learned in this new space through a "large margin" optimization process. The efficiency of the learned metric is compared to the major alternative metrics on large public datasets.

Type de document :

Communication dans un congrès

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Database (ECML-PKDD 2015), Sep 2015, Porto, Portugal. Proceedings of the ECML/PKDD 2015

Domaine :

Statistiques [stat] / Machine Learning [stat.ML]

Informatique [cs] / Intelligence artificielle [cs.AI]

Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01208429
Contributeur : Ahlame Douzal <>
Soumis le : vendredi 2 octobre 2015 - 15:34:39
Dernière modification le : jeudi 11 octobre 2018 - 08:48:02

Temporal and Frequential Metric Learning for Time Series kNN Classification

Identifiants

Collections

Citation

Exporter

Métriques

528

Contact

Temporal and Frequential Metric Learning for Time Series kNN Classification

Identifiants

Collections

Citation

Exporter

Partager

Métriques

528

Contact