Deep Distance Measurement Method for Unsupervised Multivariate Time Series Similarity Retrieval

TL;DR

This paper introduces DDMM, a novel deep learning approach for unsupervised multivariate time series similarity retrieval that effectively captures minute differences within states, improving accuracy especially in industrial applications.

Contribution

The paper presents DDMM, a new learning algorithm that assigns weights to pairs of time series segments, enabling precise similarity measurement and outperforming existing methods.

Findings

DDMM significantly outperforms state-of-the-art methods on industrial datasets.

Linking DDMM with feature extraction further improves accuracy.

Effective in recognizing subtle differences in multivariate time series.

Abstract

We propose the Deep Distance Measurement Method (DDMM) to improve retrieval accuracy in unsupervised multivariate time series similarity retrieval. DDMM enables learning of minute differences within states in the entire time series and thereby recognition of minute differences between states, which are of interest to users in industrial plants. To achieve this, DDMM uses a learning algorithm that assigns a weight to each pair of an anchor and a positive sample, arbitrarily sampled from the entire time series, based on the Euclidean distance within the pair and learns the differences within the pairs weighted by the weights. This algorithm allows both learning minute differences within states and sampling pairs from the entire time series. Our empirical studies showed that DDMM significantly outperformed state-of-the-art time series representation learning methods on the Pulp-and-paper mill dataset and demonstrated the effectiveness of DDMM in industrial plants. Furthermore, we showed that accuracy can be further improved by linking DDMM with existing feature extraction methods through experiments with the combined model.