English

Unsupervised Feature Construction for Anomaly Detection in Time Series -- An Evaluation

Machine Learning 2025-01-22 v2

Abstract

To detect anomalies with precision and without prior knowledge in time series, is it better to build a detector from the initial temporal representation, or to compute a new (tabular) representation using an existing automatic variable construction library? In this article, we address this question by conducting an in-depth experimental study for two popular detectors (Isolation Forest and Local Outlier Factor). The obtained results, for 5 different datasets, show that the new representation, computed using the tsfresh library, allows Isolation Forest to significantly improve its performance.

Cite

@article{arxiv.2501.07999,
  title  = {Unsupervised Feature Construction for Anomaly Detection in Time Series -- An Evaluation},
  author = {Marine Hamon and Vincent Lemaire and Nour Eddine Yassine Nair-Benrekia and Samuel Berlemont and Julien Cumin},
  journal= {arXiv preprint arXiv:2501.07999},
  year   = {2025}
}

Comments

7

R2 v1 2026-06-28T21:05:44.159Z