Isaacburmingham/multivariate-time-series-anomaly-detection

Analyzing multiple multivariate time series datasets and using LSTMs and Nonparametric Dynamic Thresholding to detect anomalies across various industries.

29
/ 100
Experimental

This tool helps operations managers, data analysts, or engineers overseeing complex systems identify unusual patterns in their operational data. By analyzing multiple related data streams over time—like sensor readings, network traffic, or financial metrics—it flags points where system behavior deviates significantly from normal, providing early warnings of potential issues. It takes in historical time-series data from various sources and outputs specific timestamps and data points indicating anomalies.

No commits in the last 6 months.

Use this if you need to automatically detect unexpected events or performance issues in systems that generate a lot of interconnected time-series data.

Not ideal if your data isn't time-series based, you only have a single data stream, or you need to predict future values rather than identify past anomalies.

operations-monitoring system-health predictive-maintenance fraud-detection industrial-iot
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 15 / 25

How are scores calculated?

Stars

21

Forks

5

Language

Jupyter Notebook

License

Last pushed

Jul 12, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Isaacburmingham/multivariate-time-series-anomaly-detection"

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