Iterations: 0

An Isolation Forest identifies outliers in a dataset by randomly dividing a space until each point is isolated from each other. By repeating this proceedure several times, the Isolation Forest identifies points that are quickly isolated from other points as outliers. The illustration above attempts to illustrate the method by which these outliers are quickly identified. Isolated points are colored and labelled with the iteration on which they were isolated. If you repeat the procedure several times, you'll see the outlier is consistently isolated very quickly, which allows the Isolation Forest to identify the point as an outlier.