Contextual Performance Profiling and Anomaly Detection

Description: 

This toolbox is an easy-to-use Python package for time series anomaly detection. Its main goal is to offer a simple interface to detect anomalies in time series data using state-of-the-art methods. Additionally, dtaianomaly provides a benchmark framework to quantitatively evaluate time series anomaly detection methods on a large repository of time series.

Starter Kit 1.2.2: Contextual Performance Profiling and Anomaly Detection (link: elucidata_starterkit_1_2_2.ipynb - Colab)

Documentation – Source code

More information:

• Enrichment project: An industry compliant methodology for contextual performance profiling and anomaly detection

• Responsible Research Lead: Alessandro Murgia (Sirris), Elena Tsiporkova (Sirris), Mathias Verbeke (KU Leuven) and Wannes Meert (KU Leuven)

• Authors: Louis Carpentier, Cigde Goktas (KU Leuven), Jelle Van De Sijpe, Nick Seeuws (Leuven.AI)

• Programming Language: Pyhon

•  Link to the repository: GitHub - ML-KULeuven/dtaianomaly: A simple to use Python tool for time series anomaly detection

• Categories: Time-series, anomaly detection
  

Publications:

5 min pitch on Flanders AI Research Day