Anomaly Detection with Python and Scikit Learn - All Models Crash Course!

Anomaly Detection is one of the most common use cases of Data Science and Machine Learning.

In this video we use Python and Scikit Learn to review the 2 main types of Anomaly Detection applications: Outlier Detection and Novelty Detection

We then overview the main Machine Learning models used in Scikit Learn for Anomaly detection:
Elliptic Envelope / Robust Covariance, Isolation Forests, Local Outlier Factor, and One Class SVMs

Contents

00:00 What is Anomaly Detection?
03:40 Main algorithms and models for Anomaly Detection
08:25 Novelty Detection with Scikit Learn and One-Class SVM
11:14 Novelty Detection with Local Outlier Factor
12:08 Outlier Detection with Scikit Learn
12:40 Elliptic Envelope | Robust Covariance
14:14 Isolation Forest for Anomaly Detection
18:45 Local Outlier Factor for Outlier Detection