DecisionTree

This project is about learning from decision tree, using a simple pruning technique.

How to use:

Use all the files in the repository and run the main.py. Uncomment the code releated to the dataset you want to test.

Repository Overview:

The repository has 6 files:

• dataset.py: class to represent a given dataset
• decisiontree.py: class to represent the decision tree
• decisionleaf.py: class to define a leaf of the decision tree
• decisiontreelearner.py: this file include the function to create the decision tree and to use the pruning technique
• test.py: this file contains all the functions used to create the dataset from a file that contains dataset and to test the results
• main.py: file to make tests and generate plot

Dataset:

You can download the 3 datasets used from the following links:

• car-evaluation: https://archive.ics.uci.edu/ml/datasets/Car+Evaluation
• balance-scale: https://archive.ics.uci.edu/ml/datasets/Balance+Scale
• tic-tac-toe: https://archive.ics.uci.edu/ml/datasets/Tic-Tac-Toe+Endgame

Requirements:

All the code is written in python 3

Dependencies

• math
• copy
• deepcopy
• datetime
• random
• matplotlib
• pyplot

References:

The organization of some class and part of the code is taken from : https://github.com/aimacode/aima-python