onerachel / unsupervised_anomaly_detection

Unsupervised Anomaly Detection Using Isolation Forest

Purpose

Detect abnormal samples from unlabelled data sets in three scenarios (Fraud, breast cancer, general dataset mixing categorical and numerical features).

Installation

• clone the repository

git clone [email protected]:onerachel/Unsupervised_Anomaly_Detection.git

• Create a Python virtual environment under the 'Unsupervised_Anomaly_Detection' root directory:

virtualenv -p=python3.9 .venv
source .venv/bin/activate
pip install pycaret==2.3.5
pip install scipy==1.4.1

Example of Results

Plotting