We have already seen what a decision tree is. Having understood decision trees, let's take a look at random forests. A random forest combines many decision trees into a single model. Individually, predictions made by decision trees (or humans) may not be accurate, but combined together, the predictions will be closer to the mark, on average.

The following diagram shows us a random forest, where there are multiple trees and each is making a prediction:

Random forest is a combination of many decision trees and, hence, there is a greater probability of having many views from all trees in the forest to arrive at the final desired outcome/prediction. If only a single decision tree is taken into consideration for prediction, there is less information considered for prediction. But in random forest, when there are many trees involved, the source of information is diverse and extensive. Unlike decision trees, random forests are not biased, since they are not dependent on one source. 

The following diagram demonstrates the concept of random forests:

Random forests can be applied to the following areas:

• Risk identification
• Loan processing
• Election result prediction
• Process optimization
• Optional pricing