Steering the LoCoMotif: Using Domain Knowledge in Time Series Motif Discovery
This work addresses the need for more user-relevant motif discovery in time series analysis, though it appears incremental as it builds on existing TSMD methods by adding constraint support.
The paper tackles the problem of unsupervised time series motif discovery often yielding uninteresting motifs by proposing a framework that allows users to impose domain-specific constraints, resulting in the LoCoMotif-DoK algorithm that effectively leverages domain knowledge and outperforms other techniques.
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.