Outlier Detection for a 2D Feature Space in Python

This project contains all files related to the Medium article "Outlier Detection for a 2D Feature Space in Python: How to detect outliers using plotting and clustering techniques to analyze the dependency of two features".

Motivation

This project is showing some basics steps in how to visualize and detect outliers in a two-dimensional feature space, meaning a dependency between two features.

Folder structure

• ./datasets: Contains the input dataset for the used algorithms and the outcome dataset of the outlier detection.
• ./algorithms : Contains the algorithms to visualize and detect outliers in a 2D feature space, including an unsupervised machine learning model to cluster data.

Environment

The project was developed in the IDE PyCharm with the Project Interpreter Python 3.7.

Used frameworks include

• pandas
• NumPy
• scikit-learn
• matplotlib