Random Forests are known as ensembles of decision trees, formed by combining many decision trees. Like decision trees, random forests can handle categorical features, support multiclass classification, and don't require feature scaling.
Spark ML supports random forests for both binary and multiclass classification and regression using both continuous and categorical features.
Let's train the sample lib SVM data by splitting it into 80% training and 20% testing, use Random Forest Classifier from Spark to build the model, and get evaluation metrics around the test data. The model can be persisted and loaded for later use.
Let's train the StumbleUpon dataset by splitting it into 80% training and 20% testing, use Random Forest ...