LaunchHacks-Hackaton-ML

Project: Gaming Review Analysis

Team Members:

Atila Costa Almeida (@rgenge), Shamairra (@ShamZoz), Brandon Bayquen (@bayquen)

Description:

• Python Machine-Learning project for the LaunchHacks Hackathon, July 2023.
• Understand the positive and negative sentiments of video game reviews from MetaCritic.com!
• This consists of a Metacritic game-reviews sentiment analyzer

Example Output: Word Cloud Showing Most Common Words in Cyberpunk 2077 Reviews:

• The program is divided into 2 parts: Web-Scrapping and Sentiment Analysis

The Web Scrapping involves:

1. First, inserting the path of a given Metacritic review URL for a video game (e.g. "game/pc/cyberpunk-2077")
2. Web-scrapping that URL's HTML pages to retrieve all available USER REVIEWS
3. Converting the reviews dataset into a .csv file

The Sentiment Analysis involves:

1. Using the game-reviews dataset and cleaning its data for useability
2. Splitting the dataset into two subsets: "positive" and "negative" reviews
3. Analyzing the file using data processing (i.e. Pandas, etc.) and machine-learning libraries (i.e. Wordcloud, etc.) to find the most repeated words in the datasets
4. Finally, generating a corresponding "word cloud" image for each split dataset that showcases the most used words in the reviews.

GUIDE: How To Use Our Program:

1. Please clone this repo locally into your machine in your desired directory.
2. Now, you must have Python 3 (v. 3.8 stable build or higher recommended) and Jupyter Notebook installed on your machine:

• LINUX:

sudo apt install jupyter-notebook
sudo apt install python3

• WINDOWS:

pip install jupyter-notebook
# [To install Python for Windows, go to https://www.python.org/downloads/]

• MAC:

pip install jupyter notebook
# [To install Python for Mac, go to https://www.python.org/downloads/]

3. After running make, you have to install Python package requirements using a command line (e.g. Your machine's terminal, Anaconda, etc):

• To install the package requirements and their dependencies for your machine:

• NOTE: If you're on LINUX (or using Ubuntu 18.04 or higher), please make sure to INSTALL pip first:

sudo apt update
sudo apt install python3-pip

Then confirm that pip was installed correctly by checking the pip version:

pip3 --version

4. To install the required packages and dependencies through a terminal:

pip install -r requirements.txt

• OR, using Jupyter, install the requirements inside a Jupyter Notebook kernel by opening first_time_installer.ipynb and then running it.

5. Type the following on your terminal to run the server.py source code:

python3 server.py

6. To use the sentiment analyzer:

• First, open the page 127.0.0.1:5000 (as shown in example screenshot below) in your browser:

• Next, choose a video game review page (e.g. the example paths shown in the HTML page) in the first submission box:

• Now, wait a moment as the review data is extracted and saved as a .csv file into the same directory you used in your machine.
• Once finished, it should notify you.
• Then, using Jupyter notebook on your machine, run the sentiment.ipynb Notebook file using kernel.
• The word clouds for "positive" and "negative" reviews will be outputted, assuming you installed all required Python packages correctly.
• Voilà! You're done!