4.2 Forest Size

As explained in Chapter 2 , BigML anomalies use the Isolation Forest algorithm which is an ensemble of decision trees to detect anomalies. The Forest size parameter allows you to configure the number of decision trees composing the ensemble. This must be a number up to 256 (1,000 if you are using the BigML API). Higher numbers tend to give better results although they take longer to process. By default the number of models is 128.