Random Isolation Forest

Here you’ll find everything about random isolation forests, whether it’s tutorials on implementing them in Python or conceptual articles.

A Random Isolation Forest is an ensemble machine learning model that serves the purpose of detecting anomalies in a dataset. It is an unsupervised learning algorithm. A random isolation forest works by randomly selecting a feature and a split value for that feature to create a decision tree. The decision tree is then trained on a sample of the data, and the length of the path from the root node to the leaf node that contains a given data point is calculated. In multiple iterations, this process creates a forest of decision trees. Anomalous points will have shorter path lengths on average. As a result, the algorithm can identify them by looking for points with unusually short path lengths.