A Multi-modal Metric Learning Framework for Time Series kNN Classification

Cao-Tri Do (1, 2) , Ahlame Douzal-Chouakria (3) , Sylvain Marié (4) , Michèle Rombaut (2)

1 LIG - Laboratoire d'Informatique de Grenoble
2 GIPSA-AGPIG - GIPSA - Architecture, Géométrie, Perception, Images, Gestes
3 AMA - Analyse de données, Modélisation et Apprentissage automatique [Grenoble]
4 SE - Schneider Electric

This work proposes a temporal and frequential metric learning framework for a time series nearest neighbor classification. For that, time series are embedded into a pairwise space where a combination function is learned based on a maximum margin optimization process. A wide range of experiments are conducted to evaluate the ability of the learned metric on time series kNN classification.

Mots clés: Metric learning; Time series; k-NN; Classification; Spectral metric

Domaines: Machine Learning [stat.ML]; Intelligence artificielle [cs.AI]

Cao-Tri Do, Ahlame Douzal-Chouakria, Sylvain Marié, Michèle Rombaut. A Multi-modal Metric Learning Framework for Time Series kNN Classification. Advanced Analysis and Learning on Temporal Data, pp.131-143, 2016, ⟨10.1007/978-3-319-44412-3_9⟩. ⟨hal-01385071⟩