From Time Series to Euclidean Spaces: On Spatial Transformations for Temporal Clustering

TL;DR

This paper introduces a novel temporal clustering method that transforms time series into domain-aware representations, enabling more accurate clustering despite challenges like varying sampling rates and high dimensionality.

Contribution

The authors propose a new approach combining similarity-based projections and a CNN-GRU autoencoder for effective temporal clustering, outperforming existing methods.

Findings

Outperforms existing methods by up to 32% in accuracy.

Robust across various datasets and domains.

Computationally efficient with negligible overhead.

Abstract

Unsupervised clustering of temporal data is both challenging and crucial in machine learning. In this paper, we show that neither traditional clustering methods, time series specific or even deep learning-based alternatives generalise well when both varying sampling rates and high dimensionality are present in the input data. We propose a novel approach to temporal clustering, in which we (1) transform the input time series into a distance-based projected representation by using similarity measures suitable for dealing with temporal data,(2) feed these projections into a multi-layer CNN-GRU autoencoder to generate meaningful domain-aware latent representations, which ultimately (3) allow for a natural separation of clusters beneficial for most important traditional clustering algorithms. We evaluate our approach on time series datasets from various domains and show that it not only outperforms existing methods in all cases, by up to 32%, but is also robust and incurs negligible computation overheads.