Segment and combine approach for non-parametric time-series classification

Authors:
Pierre Geurts;Louis Wehenkel
Affiliations:
Department of Electrical Engineering and Computer Science, University of Liège, Liège, Belgium;Department of Electrical Engineering and Computer Science, University of Liège, Liège, Belgium
Venue:
PKDD'05 Proceedings of the 9th European conference on Principles and Practice of Knowledge Discovery in Databases
Year:
2005

Citing 7
Cited 6

Multidimensional curve classification using passing—through regions

Pattern Recognition Letters - Special issue on pattern recognition in practice VI
Learning Comprehensible Descriptions of Multivariate Time Series

ICML '99 Proceedings of the Sixteenth International Conference on Machine Learning
Pattern Extraction for Time Series Classification

PKDD '01 Proceedings of the 5th European Conference on Principles of Data Mining and Knowledge Discovery
Generalized feature extraction for structural pattern recognition in time-series data

Generalized feature extraction for structural pattern recognition in time-series data
Automatic Feature Extraction for Classifying Audio Data

Machine Learning
Classification of Multivariate Time Series and Structured Data Using Constructive Induction

Machine Learning
Random Subwindows for Robust Image Classification

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01

Extremely randomized trees

Machine Learning
Classification of multivariate time series using two-dimensional singular value decomposition

Knowledge-Based Systems
Classification of multivariate time series using locality preserving projections

Knowledge-Based Systems
Classification trees for time series

Pattern Recognition
Decision forest: an algorithm for classifying multivariate time series

International Journal of Business Intelligence and Data Mining
The influence of global constraints on similarity measures for time-series databases

Knowledge-Based Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel, generic, scalable, autonomous, and flexible supervised learning algorithm for the classification of multi-variate and variable length time series. The essential ingredients of the algorithm are randomization, segmentation of time-series, decision tree ensemble based learning of subseries classifiers, combination of subseries classification by voting, and cross-validation based temporal resolution adaptation. Experiments are carried out with this method on 10 synthetic and real-world datasets. They highlight the good behavior of the algorithm on a large diversity of problems. Our results are also highly competitive with existing approaches from the literature.