Random Forest is used as a classifier.

In this scores are calculated using Categorization Accuracy.

Scoring Metric will expect the Solution and Submission CSVs to have certain columns, defined in a header row. For example, Accuracy expects an Id column identifying the row and an Expected column specifying the true value which competitors are attempting to predict. Your Solution CSV's header row does not need to use the exact column names, you will be able to map your column names onto the Metric's expected columns.