YouTube Live Chat Sentiment Analyser

This project is a real-time sentiment analysis tool for YouTube Live Chats, built with a Python FastAPI backend and a custom-trained machine learning model, and a TypeScript-based Chrome extension for the front end. It is designed to provide content creators and moderators with instant insights into the sentiment of their live chat audience.

---

Features

• Real-time Analysis: Monitors and analyzes YouTube live chat messages as they are posted.
• Interactive Dashboard: A web-based interface provides a dashboard with charts and key statistics for sentiment distribution.
• Confidence Scores: Each sentiment prediction includes a confidence score, giving users a better understanding of the model's certainty.
• Auto-start: The analysis automatically begins when a user navigates to a YouTube live stream page.
• Customizable: The Chrome extension is configurable, allowing users to specify the API server URL and other analysis settings.

---

Project Structure

The project is divided into two main components: a server and a chrome_extension.

Server

The server directory contains the backend of the application.

• Framework: Built using FastAPI for high performance and easy API creation.
• Machine Learning: Hosts a custom-trained sentiment analysis model. The model is trained and managed using MLflow for experiment tracking, versioning, and easy deployment.
• Tokenizer: A custom tokenizer is used to preprocess text data for the model.

Chrome Extension

The chrome_extension directory holds the user-facing part of the application.

• Languages: Developed using TypeScript to ensure a robust and maintainable codebase.
• Functionality: It injects a content script into YouTube live stream pages to capture chat messages and send them to the FastAPI server for analysis. The results are then displayed on an interactive dashboard.

---

Getting Started

Prerequisites

• Python 3.8+
• Node.js and npm
• Google Chrome browser
• MLflow (pip install mlflow)

---

Installation

1. Backend Server Setup

Navigate to the server directory and install the required Python dependencies:

pip install -r requirements.txt

2. Chrome Extension Setup

Navigate to the chrome_extension directory and install the Node.js packages:

npm install

After installation, build the extension:

npm run build

This will create a dist directory.

---

Running the Application

Start the Server

From the server directory, run the FastAPI application:

uvicorn main:app --reload

The server will start at http://127.0.0.1:8000.

Load the Chrome Extension

1. Open Google Chrome and navigate to chrome://extensions/.
2. Enable "Developer mode" in the top-right corner.
3. Click "Load unpacked" and select the dist folder from your chrome_extension directory.
4. The extension icon will appear in your browser toolbar.

Start Analyzing

• Go to a YouTube live stream with an active chat.
• Click the extension icon to open the popup.
• The analysis should begin automatically.

---

MLflow

The MLflow tracking server is used to manage and serve the machine learning models.

To run the MLflow UI and view your experiments:

1. Navigate to the project's root directory.
2. Start the MLflow UI:

mlflow server --host 127.0.0.1 --port 5000

You can then access the UI in your browser at: 👉 http://127.0.0.1:5000

---

License

This project is licensed under the MIT License. See the LICENSE file for details.