Google Reviews Analyzer Chrome Extension

A Chrome extension that helps analyze Google Reviews to determine their authenticity using machine learning and sentiment analysis.

Features

• Real-time review analysis
• Fake review detection
• Sentiment analysis
• Detailed reasoning for predictions
• Report suspicious reviews
• Continuous model improvement through user feedback

Tech Stack

• Frontend: Chrome Extension (JavaScript)
• Backend: Flask (Python)
• Machine Learning:
  • NLTK for sentiment analysis
  • Custom classifier for fake review detection
• Styling: TailwindCSS
• Build Tool: Webpack

Installation

1. Clone the repository:

git clone [repository-url]

2. Install Python dependencies:

pip install -r requirements.txt

3. Install Node.js dependencies:

npm install

4. Build the extension:

npm run build

5. Load the extension in Chrome:
   • Open Chrome and navigate to chrome://extensions/
   • Enable "Developer mode"
   • Click "Load unpacked"
   • Select the dist directory in your project folder

Development

Backend Setup

1. Start the Flask server:

python app.py

Extension Development

1. Make changes to files in the src directory
2. Rebuild the extension:

npm run build

3. Reload the extension in Chrome

Project Structure

├── extension/
│   └── src/
│       ├── externalHTML/
│       │   ├── error.js
│       │   └── feedback.js
│       └── reviews/
│           ├── background.js
│           ├── content.js
│           └── popup.js
├── ml/
│   ├── model.py
│   └── train_model.py
├── app.py
└── requirements.txt

How It Works

1. The extension monitors Google Review pages
2. When a review is detected, it's sent to the Flask backend
3. The backend processes the review using:
   • NLTK sentiment analysis
   • Custom trained classifier
4. Results are displayed in the extension popup
5. Users can report suspicious reviews
6. The model is periodically retrained with user feedback

Acknowledgments

• NLTK for sentiment analysis
• TailwindCSS for styling
• Webpack for building