Randomized trees for time series representation and similarity. (December 2021)

• Record Type: Journal Article
• Authors: Görgülü, Berk; Baydoğan, Mustafa Gökçe
• Is Part Of: Pattern recognition. Volume 120(2021)
• Publication Date: 2021-12
• Subjects: Time series -- Representation learning -- Random trees -- Classification; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2021.108097
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Representation learning is difficult when time series contain irregularities. Rand-TS is a time series representation learning framework based on random trees. Rand-TS can work with both univariate and multivariate time series. Rand-TS can work with time series with varying length and missing information. Allows incorporating additional information into the time series representation. Abstract: Most of the temporal data mining tasks require representations to capture important characteristics of time series. Representation learning is challenging when time series differ in distributional characteristics and/or show irregularities such as varying lengths and missing observations. Moreover, when time series are multivariate, interactions between variables should be modeled efficiently. This study proposes a unified, flexible time series representation learning framework for both univariate and multivariate time series called Rand-TS. Rand-TS models density characteristics of each time series as a time-varying Gaussian distribution using random decision trees and embeds density information into a sparse vector. Rand-TS can work with time series of various lengths and missing observations, furthermore, it allows using customized features. We illustrate the classification performance of Rand-TS on 113 univariate, 19 multivariate along with 15 univariate time series with varying lengths from UCR database. The results show that in addition to its flexibility, Rand-TS provides competitive classification performance.