boniolp/MSAD

[VLDB 2023] Model Selection for Anomaly Detection in Time Series

47
/ 100
Emerging

This project helps data scientists, machine learning engineers, and researchers working with time series data to choose the best anomaly detection model for their specific dataset. It takes raw time series and their associated anomaly scores as input and outputs a recommendation for the most effective anomaly detection method. The end-user is anyone needing to reliably identify unusual patterns in their time-series data.

Use this if you need to select the most effective anomaly detection model for your time series data and want to avoid trial-and-error, ensuring optimal performance.

Not ideal if you are looking for new anomaly detection algorithms themselves, rather than methods for selecting existing ones.

time-series-analysis anomaly-detection model-selection predictive-maintenance fraud-detection
No Package No Dependents
Maintenance 6 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

43

Forks

10

Language

Jupyter Notebook

License

MIT

Last pushed

Dec 02, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/boniolp/MSAD"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.