datasig-ac-uk/signature_mahalanobis_knn
Methodology for anomaly detection on multivariate streams using path signatures and the variance norm.
This tool helps you automatically identify unusual behavior or events within complex, evolving datasets, such as sensor readings over time or financial transaction logs. It takes a history of normal data streams (each stream having multiple changing measurements) as input, and then tells you how 'anomalous' new streams are compared to that normal behavior. Data scientists, operations engineers, or fraud analysts who need to monitor systems for unexpected deviations would find this useful.
No commits in the last 6 months. Available on PyPI.
Use this if you need to detect subtle anomalies in multivariate time series data where the relationships between different measurements are important and change over time.
Not ideal if your data is static, not streamed, or if you're looking for simple threshold-based anomalies rather than complex pattern deviations.
Stars
9
Forks
3
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Dec 09, 2024
Commits (30d)
0
Dependencies
7
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/datasig-ac-uk/signature_mahalanobis_knn"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
yzhao062/pyod
A Python Library for Outlier and Anomaly Detection, Integrating Classical and Deep Learning Techniques
unit8co/darts
A python library for user-friendly forecasting and anomaly detection on time series.
elki-project/elki
ELKI Data Mining Toolkit
raphaelvallat/antropy
AntroPy: entropy and complexity of (EEG) time-series in Python
Minqi824/ADBench
Official Implement of "ADBench: Anomaly Detection Benchmark", NeurIPS 2022.