Piforest 🌳🌲🌳🌲

Implementation of the anomaly detection in resource constrained environments with streaming data by jain et al. (2021).

Jain, Prarthi, Seemandhar Jain, Osmar R. Zaïane, and Abhishek Srivastava. "Anomaly Detection in Resource Constrained Environments With Streaming Data." IEEE Transactions on Emerging Topics in Computational Intelligence (2021).

About

The Preprocessed isolation forest (PiForest) algorithm is a method for detecting outliers in streaming data for resource constrained environment. PiForest offers a number of features that many competing anomaly detection algorithms lack. Specifically, PiForest:

• Is designed to handle streaming data.
• Performs well in resource constrained environment.
• Performs well on high-dimensional data.
• Reduces the influence of irrelevant dimensions.
• Features an anomaly-scoring algorithm with a clear underlying statistical meaning.

This repository provides an open-source implementation of the PiForest algorithm and its core data structures for the purposes of facilitating experimentation and enabling future extensions of the PiForest algorithm.

Documentation

Read the docs here 📖.

Installation

Download the repo, and run the main file. Currently, only Python 3 is supported.

Dependencies

The following dependencies are required to install and use PiForest:

-numpy==1.19.2 -pandas==1.1.3 -scikit_multiflow==0.5.3 -scikit_learn==0.24.2

Listed version numbers have been tested and are known to work (this does not necessarily preclude older versions).

Citing

If you have used this codebase in a publication and wish to cite it, please use the IEEE Transactions on Emerging Topics in Computational Intelligence.

Jain, Prarthi, Seemandhar Jain, Osmar R. Zaïane, and Abhishek Srivastava. "Anomaly Detection in Resource Constrained Environments With Streaming Data." IEEE Transactions on Emerging Topics in Computational Intelligence (2021).

@article{jain2021anomaly,
  title={Anomaly Detection in Resource Constrained Environments With Streaming Data},
  author={Jain, Prarthi and Jain, Seemandhar and Za{\"\i}ane, Osmar R and Srivastava, Abhishek},
  journal={IEEE Transactions on Emerging Topics in Computational Intelligence},
  year={2021},
  publisher={IEEE}
}

PiForest Working

The authors present the Preprocessed Isolation Forest PiForest approach for anomaly detection that works well in resource constrained environments and is also effective on streaming data. Propsed Architecture

Cicular Queue

Arduino Real world Set-up