MINT: Tensor Decomposition on Stacked Recurrence Matrices for Time Series Data Mining

2026-08-04

AI Summary

This study addresses the challenge of efficiently discovering cross-series and cross-temporal co-occurring patterns and periodic structures in large-scale univariate and multivariate time series. To this end, the authors propose a tensorized recurrence plot representation that naturally extends to multivariate settings, transforming time series into a three-dimensional tensor of stacked self-similarity matrices of size N×(n−m+1)×(n−m+1). They further develop an end-to-end MINT pipeline that integrates tensor decomposition to jointly cluster periodic patterns across sensors. Experiments on real-world datasets—including subway passenger flow, power load, wind power generation, and traffic volume—demonstrate that the method effectively identifies highly regular and sensor-consistent periodic co-occurrence patterns.

Abstract

Recurrence plots are a time series data mining primitive applied to a variety of domains (e.g. star light curves, sound waveforms, CCT telemetry). This work proposes tensorized self-similarity matrices as a primitive for univariate time series datasets ($N\times n$) of $N$ time series of length $n$ with a subsequence window of length $m$, and whose tensor-based nature is naturally extensible to multivariate datasets. The proposed method to compute this primitive computes dot plots of size $N \times (n-m+1) \times (n-m+ 1)$ from these datasets, where the subsequent tensor is mined using tensor decomposition methods to mine for co-clustered patterns. We demonstrate our results in mass rapid transit, electricity demand, wind turbine, and car traffic data, finding the MINT pipeline effectively co-clusters cross-sensor patterns in highly regular datasets containing motifs at regular intervals.

Problem

• time series
• tensor decomposition
• recurrence plots
• co-clustering
• motif discovery

Innovation

• tensor decomposition
• recurrence plots
• time series co-clustering
• self-similarity matrices
• multivariate time series