Dota2 Team Winner Prediction

This repository contains the project for Machine Learning course 2024/2025 and it is structured as follows:

Folder Structure

• code/: Contains all the scripts and Jupyter notebooks used for data processing, feature selection, model training, and evaluation.

  • adaboost.ipynb: Implementation of the AdaBoost model.
  • data.ipynb: Information about the dataframes and their columns.
  • dataset_functions.py: Utility functions for dataset handling, mainly transformations and feature engineering.
  • feature_importance.ipynb: Analysis of feature importance.
  • feature_selection.ipynb: Feature selection methods.
  • final_predictions.ipynb: Final model predictions.
  • gradboost.ipynb: Implementation of the Gradient Boosting model.
  • knn.ipynb: Implementation of the k-Nearest Neighbors (k-NN) model.
  • random_forest.ipynb: Implementation of the Random Forest model.
  • SVM.ipynb: Implementation of the Support Vector Machine (SVM) model.
  • visualization.ipynb: Data visualization and exploratory analysis.

• dataset/: Contains the dataset used in the project.

• doc/: Includes documentation related to the project.

• submission.csv: The final predictions generated by the model.

Usage

To run this project, ensure you have all dependencies installed. Open the Jupyter notebooks inside the code/ folder to explore data, train models, and generate predictions.

Authors

Luca Panariello & Enrico Loda