Published February 3, 2026
Library
graduated
Topological Data Analysis for Anomaly Detection
PythonPackage
Maintainer:IRT-SystemX
Description
tdaad provides machine learning algorithms for analyzing timeseries data through the lense of Topological Data Analysis, and deriving anomaly scores.
The targeted input is an object X representing a multiple time series with variables columns and timestamps lines. We use the term multiple time series to describe a set of univariate timeseries that describe a system or object. Note that the package does not handle analysis of a single univariate timeseries.
The main idea of this package is to analyze time series with topological methods.
Owner:IRT-SystemX
Keywords:tdaad
CONTEXT
Anomaly detection is a major challenge in artificial intelligence, as it enables
the automatic identification of unusual or suspicious patterns in data, which is
essential both for ensuring the reliability of systems and for preventing risks in
critical domains such as cybersecurity, healthcare, or industrial maintenance
VALUE PROPOSITION
This library proposes an anomaly detection method on multivariate timeseries based on topological analysis. As it is a kind of unsupervised method, it does not require labelled data to be functional.
WHEN TO USE IT
It shall be used both during the training and the production step to monitor data.
RESOURCES
