Steering the LoCoMotif: Using Domain Knowledge in Time Series Motif Discovery

2025-02-17

AI Summary

Time-series motif discovery (TSMD) is inherently unsupervised and often yields motifs misaligned with users’ semantic intent. Method: We propose the first plug-and-play domain-knowledge injection framework, enabling users to flexibly define, combine, and extend diverse constraints—such as shape, phase, and physical range—based on prior domain knowledge. To realize this, we design LoCoMotif-DoK, a constraint-aware motif search algorithm that integrates dynamic programming with aggressive pruning, achieving sub-quadratic time complexity while preserving expressive power and computational efficiency. Contribution/Results: Extensive experiments on real-world and synthetic datasets demonstrate that our approach significantly outperforms state-of-the-art TSMD methods, improving recall by up to 41% and precisely satisfying user-specified semantic requirements.

Technology Category

• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Search and Optimization: Local Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Topic discovery and tracking

Abstract

Time Series Motif Discovery (TSMD) identifies repeating patterns in time series data, but its unsupervised nature might result in motifs that are not interesting to the user. To address this, we propose a framework that allows the user to impose constraints on the motifs to be discovered, where constraints can easily be defined according to the properties of the desired motifs in the application domain. We also propose an efficient implementation of the framework, the LoCoMotif-DoK algorithm. We demonstrate that LoCoMotif-DoK can effectively leverage domain knowledge in real and synthetic data, outperforming other TSMD techniques which only support a limited form of domain knowledge.

Problem

• Incorporates domain knowledge in motif discovery
• Improves relevance of discovered time series motifs
• Enhances efficiency with LoCoMotif-DoK algorithm

Innovation

• Domain knowledge integration
• User-defined motif constraints
• Efficient LoCoMotif-DoK algorithm