TFEC: Multivariate Time-Series Clustering via Temporal-Frequency Enhanced Contrastive Learning

2026-01-12

AI Summary

This work addresses critical limitations in existing contrastive learning–based approaches for multivariate time series clustering, which often neglect inherent cluster structures when constructing positive and negative samples and employ data augmentation strategies that disrupt temporal dependencies and periodicity, thereby degrading representation quality. To overcome these issues, the authors propose TFEC, a novel framework featuring a time–frequency collaborative augmentation mechanism that preserves essential temporal structures while avoiding high-distortion perturbations. Additionally, TFEC introduces a dual-path joint optimization strategy that simultaneously learns representations and cluster assignments, enabling cluster-aware contrastive learning. Extensive experiments on six real-world datasets demonstrate that TFEC achieves an average 4.48% improvement in Normalized Mutual Information (NMI) over state-of-the-art methods, highlighting its superior clustering performance.

Technology Category

• Machine Learning: Clustering
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Multivariate Time-Series (MTS) clustering is crucial for signal processing and data analysis. Although deep learning approaches, particularly those leveraging Contrastive Learning (CL), are prominent for MTS representation, existing CL-based models face two key limitations: 1) neglecting clustering information during positive/negative sample pair construction, and 2) introducing unreasonable inductive biases, e.g., destroying time dependence and periodicity through augmentation strategies, compromising representation quality. This paper, therefore, proposes a Temporal-Frequency Enhanced Contrastive (TFEC) learning framework. To preserve temporal structure while generating low-distortion representations, a temporal-frequency Co-EnHancement (CoEH) mechanism is introduced. Accordingly, a synergistic dual-path representation and cluster distribution learning framework is designed to jointly optimize cluster structure and representation fidelity. Experiments on six real-world benchmark datasets demonstrate TFEC's superiority, achieving 4.48% average NMI gains over SOTA methods, with ablation studies validating the design. The code of the paper is available at: https://github.com/yueliangy/TFEC.

Problem

• Multivariate Time-Series Clustering
• Contrastive Learning
• Temporal Dependence
• Periodicity
• Inductive Bias

Innovation

• Temporal-Frequency Enhancement
• Contrastive Learning
• Multivariate Time-Series Clustering
• Co-EnHancement Mechanism
• Dual-Path Learning