Time series classification for varying length series

TL;DR

This paper investigates methods for classifying time series data with varying lengths, focusing on the effects of sampling rate differences and start-end point variations, and offers practical guidelines for handling such variations.

Contribution

It identifies two key mechanisms causing length variation and evaluates strategies to improve classification accuracy for these cases.

Findings

Sampling rate variations significantly affect classification performance.

Start-end point variations require specific preprocessing techniques.

Practical recommendations improve classification robustness.

Abstract

Research into time series classification has tended to focus on the case of series of uniform length. However, it is common for real-world time series data to have unequal lengths. Differing time series lengths may arise from a number of fundamentally different mechanisms. In this work, we identify and evaluate two classes of such mechanisms -- variations in sampling rate relative to the relevant signal and variations between the start and end points of one time series relative to one another. We investigate how time series generated by each of these classes of mechanism are best addressed for time series classification. We perform extensive experiments and provide practical recommendations on how variations in length should be handled in time series classification.