Journal article
Random projection ensemble classification with high-dimensional time series
Biometrics, Vol.79(2), pp.964-974
06/2023
DOI: 10.1111/biom.13679
PMID: 35426119
Appears in UI Libraries Support Open Access
Abstract
Multivariate time-series (MTS) data are prevalent in diverse domains and often high dimensional. We propose new random projection ensemble classifiers with high-dimensional MTS. The method first applies dimension reduction in the time domain via randomly projecting the time-series variables into some low dimensional space, followed by measuring the disparity via some novel base classifier between the data and the candidate generating processes in the projected space. Our contributions are two-fold: (i) we derive optimal weighted majority voting schemes for pooling information from the base classifiers for multiclass classification, and (ii) we introduce new base frequency-domain classifiers based on Whittle likelihood (WL), Kullback-Leibler divergence (KL), Eigen-Distance (ED) and Chernoff divergence (CH). Both simulations for binary and multiclass problems, and an EEG application demonstrate the efficacy of the proposed methods in constructing accurate classifiers with high-dimensional MTS. This article is protected by copyright. All rights reserved.
Details
- Title: Subtitle
- Random projection ensemble classification with high-dimensional time series
- Creators
- Fuli Zhang - Department of Statistics and Actuarial Science, University of Iowa, Iowa City Iowa.Kung-Sik Chan - University of Iowa
- Resource Type
- Journal article
- Publication Details
- Biometrics, Vol.79(2), pp.964-974
- Publisher
- Wiley
- DOI
- 10.1111/biom.13679
- PMID
- 35426119
- eISSN
- 1541-0420
- Language
- English
- Electronic publication date
- 04/15/2022
- Date published
- 06/2023
- Academic Unit
- Statistics and Actuarial Science; Radiology
- Record Identifier
- 9984257724602771
Metrics
25 Record Views