University of Michigan - Ann Arbor

Isolation forests isolate outliers found using anomaly detection. This is adverse to other techniques that typically use data profiling (which are statistical methods used to describe the data set as a whole).


Isolation  Forest

Isolation Forest Code Example

- Finance: finding anomalies of fraudulent purchases, high bank deposit regulation
- Networking: outside threats to intrude on networks


Examples of when to use Isolation Forests


Pros:
- Do not need to scale given values
- If no given distribution fits the data, this is a good option (because you can not use statistical analysis for outliers)
- Few parameters
Cons:
- Must use Sklearn 
- Visualization of given results is more complex than traditional outlier detection methods
- Training time can be extremely long if it is implemented incorrectly


Learn Before

Related