Exploring Random Forest hyperparameters
Random Forest is a tree-based model that is built using a collection of decision trees. It is a very powerful ensemble ML model that can be utilized for both classification and regression tasks. The way Random Forest utilizes the collection of decision trees is by performing an ensemble method called bootstrap aggregation (bagging) with some modifications. To understand how each of the Random Forest’s hyperparameters can impact the model’s performance, we need to understand how the model works in the first place.
Before discussing how Random Forest ensembles a collection of decision trees, let’s discuss how a decision tree works at a high level. A decision tree can be utilized to perform a classification or regression task by constructing a series of decisions (in the form of rules and splitting points) that can be visualized in the form of a tree. These decisions are made by looking through all of the features and the...