About   |   Features   |   Requirements   |   Starting   |   License   |   Author   |   Acknowledgments

Proposed work on Artificial Intelligence by professor and Ph.D Claudia Martins to implement from scratch Decision Tree and Perceptron and compare the performance between those two ML algorithms through a small dataset.

Both Decision Tree and Perceptron had 100% of accuracy on training data. Modifying parameters can demonstrate different classifications on new data and due to not having the class for this data, any modification will depends on the target domain for choose which algorithm use.

✔️ Perceptron and Decision Tree adaptable to multiple features (considering first column as index and last one the target);
✔️ Perceptron customizable;

Before starting 🏁, you need to have a compiler which supports c++17.

# Clone this project
$ git clone https://github.com/SousaPedroso/dtp

# Access
$ cd dtp

# Compile files (Considering g++, change for your compiler)
$ g++ main.cpp Node.cpp DecisionTree.cpp Perceptron.cpp error.cpp encoder.cpp utils.cpp -o main -std=c++17

# Run the project
$ .\main

# Follow the instructions in the program

This project is under license from MIT. For more details, see the LICENSE file.

Claudia Martins Mauro de Souza  

Back to top