🎯 YouTube Comment Classification Web App

This project is a full-stack web application that classifies YouTube video comments using machine learning. It provides a clean interface to fetch, analyze, and visualize comments for content creators and community managers.

🔍 Features

• 🎥 YouTube Comment Scraper: Fetches comments using YouTube Data API based on Channel ID.
• 🧠 ML Model Integration: Classifies comments into categories like Positive, Negative, Neutral, and Questions.
• 🌗 Dark/Light Mode: Toggle-friendly interface with theme persistence.
• 📈 Analysis Dashboard: View categorized comments and data insights on a dedicated page.
• ✍️ Visitor Feedback System: Allows users to submit comments/feedback that are dynamically rendered on the page.
• 📱 Responsive Design: Fully mobile-optimized with animated transitions and collapsible menus.
• 🔄 Dynamic Image Display: Rotating images in the About section for visual enhancement.

📁 Project Structure

project/ │ ├── app.py # Flask backend logic ├── sentimentscraper.py # YouTube scraper + comment classifier ├── requirements.txt # Python dependencies │ ├── templates/ │ ├── index.html # Homepage with input form and feedback section │ └── about.html # View analysis & dynamic comment section │ ├── static/ │ ├── style.css # Global styling │ ├── script1.js # JS logic (theme toggle, API calls, feedback system) │ └── images/, videos/ # Visual assets

🧠 Tech Stack

• Frontend: HTML5, CSS3, JavaScript, Font Awesome
• Backend: Python 3, Flask
• APIs & Libraries:
  • YouTube Data API
  • Custom ML classification logic (classify_text())
  • google-api-python-client
  • flask, requests, etc.

💡 Use Cases

• 📊 Analyze audience sentiment on YouTube
• 👮‍♂️ Flag potential spam or hate comments
• 💬 Collect and showcase community feedback
• 👨‍🏫 Demonstrate full-stack ML web integration

🛠 How It Works

1. User enters a valid YouTube Channel ID.
2. App fetches recent video comments using the YouTube API.
3. Each comment is run through a machine learning classifier.
4. Results are sorted into categories and displayed visually.
5. Users can also leave feedback, which appears dynamically.

📬 Feedback Handling

• All feedback is submitted via a form on the homepage.
• Submitted messages are stored (temporarily or permanently depending on backend setup).
• New feedback is dynamically added to the community section.

🧑‍💻 Author

Made with ❤️ by Harshit Bansal