Weighted kNN and constrained elastic distances for time-series classification

•The weighted k-nearest neighbor rule outperforms the 1-nearest neighbor classifier.•Recommendations for choosing weighting schemes.•Recommendations for choosing the constraint width r and the neighborhood size k. Time-series classification has been addressed by a plethora of machine-learning techni...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Expert systems with applications Jg. 162; S. 113829
Hauptverfasser: Geler, Zoltan, Kurbalija, Vladimir, Ivanović, Mirjana, Radovanović, Miloš
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Elsevier Ltd 30.12.2020
Elsevier BV
Schlagworte:
1-Nearest neighbor
Bayesian analysis
Classification
Classifiers
Constraints
Distance measurement
Elastic distance measures
k-Nearest neighbor
K-nearest neighbors algorithm
Machine learning
Neural networks
Support vector machines
Time series
Weighted k-nearest neighbor
Weighting
Elastic distance measures
1-Nearest neighbor
Weighted k-nearest neighbor
Time series
Classification
k-Nearest neighbor
ISSN:0957-4174, 1873-6793
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•The weighted k-nearest neighbor rule outperforms the 1-nearest neighbor classifier.•Recommendations for choosing weighting schemes.•Recommendations for choosing the constraint width r and the neighborhood size k. Time-series classification has been addressed by a plethora of machine-learning techniques, including neural networks, support vector machines, Bayesian approaches, and others. It is an accepted fact, however, that the plain vanilla 1-nearest neighbor (1NN) classifier, combined with an elastic distance measure such as Dynamic Time Warping (DTW), is competitive and often superior to more complex classification methods, including the majority-voting k-nearest neighbor (kNN) classifier. With this paper we continue our investigation of the kNN classifier on time-series data and the impact of various classic distance-based vote weighting schemes by considering constrained versions of four common elastic distance measures: DTW, Longest Common Subsequence (LCS), Edit Distance with Real Penalty (ERP), and Edit Distance on Real sequence (EDR). By performing experiments on the entire UCR Time Series Classification Archive we show that weighted kNN is able to consistently outperform 1NN. Furthermore, we provide recommendations for the choices of the constraint width parameter r, neighborhood size k, and weighting scheme, for each mentioned elastic distance measure.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2020.113829