Anomaly Detection in Time-Series Data

Using Machine Learning (Robust Random Cut Forest) and Multiprocessing

About The Project

Purpose for this tutorial is to show how to easily utilize a Robust Random Cut Forest neural network to find anomalies in time series data.
It utilizes multiprocessing to efficiently run multiple RRCF processes in parallell. This will decrease the execution time.

The code in main.py uses 'example-dataset.csv' which is a set of mocked time-series data.

Find anomalies in financial data

By executing "currency.py" it also shows how it can be adapted to detect unusual movements in financial instruments.

Getting Started

Prerequisites

• Python pip packages

  python3 -m pip install -r requirements.txt

Usage

• Run the code

  python3 main.py

Contributing

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/featureName)
3. Commit your Changes (git commit -m 'Add some featureName')
4. Push to the Branch (git push origin feature/featureName)
5. Open a Pull Request

Contact

Martin Karlsson

LinkedIn : martin-karlsson
Twitter : @HelloKarlsson
Email : hello@martinkarlsson.io
Webpage : www.martinkarlsson.io

Project Link: github.com/martinkarlssonio/timeseries-anomaly-detection