Implementation of Decision Tree and Random Forest classifiers in Python and Scala languages

Python

• Both classifiers use Python3 and don't need any third-party library
• The decision tree example can be launched by running:

cd python
./DecisionTree.py

• The random forest example cam be launched the same way:

cd python
./RandomForest.py

Scala

The scala version contains a main function running the same example using the decision tree and the random forest classifier. It can be launched like this:

cd scala
sbt run

The error rate of the two algorithms will be printed out.

Dataset

The dataset used in the examples is the "Census Income" dataset. It classifies incomes of approximately 32000 people into two categories: above and below 50,000$ a year.

This dataset has been created by Ronny Kohavi and Barry Becker (1994) and is distributed by the UCI Machine Learning Repository: http://archive.ics.uci.edu/ml. It can be found here: http://archive.ics.uci.edu/ml/datasets/Adult.