Comparison between Random Forests and Decision Trees




I read somewhere that random forests work on the principle of taking random sample of variables to form decision trees. Although decision trees overfit, using random forests averages this effect of overfitting as each of the trees in the forest overfit differently which averages out the overall effect.

Does this mean that random forests would always perform better than decision trees on any model?
Or are there situations where decision trees would perform better than random forests. If not, then do we use decision trees just to plot them and extract the information about the significance of the variables?




Random forest can show better power but not stable as compared to CART model and CART model can be easily understood by end users also.

You can refer below article for more detail:

Hope this helps!



In my experience, Random Forest almost always perform better than decision trees. I think the only advantage of decision trees is interpretability and perhaps training time (if that is an issue).


I have not seen decision trees performing better than random forests.

However, running decision trees doenst require me to clean the data, especially missing values. I can run the model quickly with out concentrating much on pre-processing. Also random forest requires lot of processing power. If you want to overcome overfit, you can also consider taking K-fold cross validation along with decision trees.