Time series classification by class-specific Mahalanobis distance measures

Authors:
Zoltán Prekopcsák;Daniel Lemire
Affiliations:
Budapest University of Technology and Economics, Budapest, Hungary 1117;LICEF, Université du Québec à Montréal (UQAM), Montreal, Canada H2X 3P2
Venue:
Advances in Data Analysis and Classification
Year:
2012

Citing 20
Cited 0

Discriminant Adaptive Nearest Neighbor Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
Semidefinite programming

SIAM Review
A Review and Empirical Evaluation of Feature Weighting Methods for aClass of Lazy Learning Algorithms

Artificial Intelligence Review - Special issue on lazy learning
A class-dependent weighted dissimilarity measure for nearest neighbor classification problems

Pattern Recognition Letters
MindReader: Querying Databases Through Multiple Examples

VLDB '98 Proceedings of the 24rd International Conference on Very Large Data Bases
Local feature extraction and its applications using a library of bases

Local feature extraction and its applications using a library of bases
An Adaptable Time Warping Distance for Time Series Learning

ICMLA '06 Proceedings of the 5th International Conference on Machine Learning and Applications
Learning prototypes and distances: A prototype reduction technique based on nearest neighbor error minimization

Pattern Recognition
Chromosome classification using dynamic time warping

Pattern Recognition Letters
Toward accurate dynamic time warping in linear time and space

Intelligent Data Analysis
Querying and mining of time series data: experimental comparison of representations and distance measures

Proceedings of the VLDB Endowment
Faster retrieval with a two-pass dynamic-time-warping lower bound

Pattern Recognition
Learning instance specific distances using metric propagation

ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning
A method of learning weighted similarity function to improve the performance of nearest neighbor

Information Sciences: an International Journal
Distance Metric Learning for Large Margin Nearest Neighbor Classification

The Journal of Machine Learning Research
Large margin nearest local mean classifier

Signal Processing
Minimum classification error training in example based speech and pattern recognition using sparse weight matrices

Journal of Computational and Applied Mathematics
Histogram distance for similarity search in large time series database

IDEAL'10 Proceedings of the 11th international conference on Intelligent data engineering and automated learning
Dynamic time warping constraint learning for large margin nearest neighbor classification

Information Sciences: an International Journal
Weighted dynamic time warping for time series classification

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

To classify time series by nearest neighbors, we need to specify or learn one or several distance measures. We consider variations of the Mahalanobis distance measures which rely on the inverse covariance matrix of the data. Unfortunately--for time series data--the covariance matrix has often low rank. To alleviate this problem we can either use a pseudoinverse, covariance shrinking or limit the matrix to its diagonal. We review these alternatives and benchmark them against competitive methods such as the related Large Margin Nearest Neighbor Classification (LMNN) and the Dynamic Time Warping (DTW) distance. As we expected, we find that the DTW is superior, but the Mahalanobis distance measures are one to two orders of magnitude faster. To get best results with Mahalanobis distance measures, we recommend learning one distance measure per class using either covariance shrinking or the diagonal approach.