mlsanityData to 14 Oct 2026

PHEATPRUNER: Interpretable Data-centric Feature Selection for Multivariate Time Series Classification through Persistent Homology

arXiv 2504.18329 ↗First posted 25 Apr 2025cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Balancing performance and interpretability in multivariate time series classification is a significant challenge due to data complexity and high dimensionality. This paper introduces PHeatPruner, a method integrating persistent homology and sheaf theory to address these challenges. Persistent homology facilitates the pruning of up to 45% of the applied variables while maintaining or enhancing the accuracy of models such as Random Forest, CatBoost, XGBoost, and LightGBM, all without depending on posterior probabilities or supervised optimization algorithms. Concurrently, sheaf theory contributes explanatory vectors that provide deeper insights into the data's structural nuances. The approach was validated using the UEA Archive and a mastitis detection dataset for dairy cows. The results demonstrate that PHeatPruner effectively preserves model accuracy. Furthermore, our results highlight PHeatPruner's key features, i.e. simplifying complex data and offering actionable insights without increasing processing time or complexity. This method bridges the gap between complexity reduction and interpretability, suggesting promising applications in various fields.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 43 references that are in the map. First posted on arXiv 25 Apr 2025.

Builds on

15 of 43 references.

References · 15

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2016RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanismcited by 349
  3. 2018The UEA multivariate time series classification archive, 2018cited by 338
  4. 2017A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Predictioncited by 245
  5. 2017An introduction to Topological Data Analysis: fundamental and practical aspects for data scientistscited by 172
  6. 2018Multivariate LSTM-FCNs for Time Series Classificationcited by 120
  7. 2017Multivariate Time Series Classification with WEASEL+MUSEcited by 53
  8. 2020XCM: An Explainable Convolutional Neural Network for Multivariate Time Series Classificationcited by 24
  9. 2019DSTP-RNN: a dual-stage two-phase attention-based recurrent neural networks for long-term and multivariate time series predictioncited by 20
  10. 2020Spatiotemporal Attention for Multivariate Time Series Prediction and Interpretationcited by 18
  11. 2022Unveiling the higher-order organization of multivariate time seriescited by 17
  12. 2015Ultra-Fast Shapelets for Time Series Classificationcited by 7
3 more
  1. 2015Contextuality, Cohomology and Paradoxcited by 4
  2. 2022TSEM: Temporally Weighted Spatiotemporal Explainable Neural Network for Multivariate Time Seriescited by 4
  3. 2013Automatic recognition and tagging of topologically different regimes in dynamical systemscited by 1

Built on by

Not cited yet.