thaoquynh0603 / game-recommendation

Game Recommendation

---

Table of Contents

1. Introduction
2. Installation
3. Usage
4. Features
5. Dependencies
6. Configuration
7. Documentation
8. Examples
9. Troubleshooting
10. Contributors
11. License

---

Introduction

Game Recommendation is a system that suggests the five most similar Nintendo Switch games to a chosen game using data scrapping, data cleaning, unsupervised learning (K-means), and natural language processing (NLP).

Installation

To install and run this project, follow these steps:

1. Clone the repository:

   git clone https://github.com/thaoquynh0603/game-recommendation.git

2. Navigate to the project directory:

   cd game-recommendation

3. Install the required dependencies:

   pip install -r requirements.txt

Usage

1. Open the Jupyter notebook game_recommendation.ipynb.
2. Run the notebook cells to generate game recommendations.

Features

• Data scrapping using BeautifulSoup4.
• Data cleaning and preprocessing.
• Unsupervised learning using K-means clustering.
• Natural language processing for game similarity recommendations.

Dependencies

• BeautifulSoup4
• pandas
• scikit-learn
• Jupyter Notebook

Configuration

No specific configuration is required for this project. Simply ensure that all dependencies are installed.

Documentation

Detailed documentation for each part of the project can be found within the Jupyter notebook game_recommendation.ipynb.

Examples

An example of generating game recommendations:

# Assuming you have already run the necessary cells for data preprocessing
chosen_game = 'The Legend of Zelda: Breath of the Wild'
recommended_games = get_recommendations(chosen_game)
print(recommended_games)

Troubleshooting

• Ensure all dependencies are installed.
• Verify that the dataset dataset.csv is present in the project directory.
• Check that Jupyter Notebook is properly installed and configured.

Contributors

• Thao Quynh

License

This project does not currently have a license.