Skip to content

vaibhavguptahere/Faktify-FakeNewsDetection

Repository files navigation

📢 Faktify - AI-Powered Fake News Detection

Faktify is an AI-powered fake news detection platform that empowers users to verify the authenticity of news articles. By leveraging Gemini API for intelligent comparison, it generates a credibility scorecard, helping users quickly determine the reliability of online content.

Our mission is to combat misinformation and promote responsible digital engagement with an interactive and user-friendly experience.


🚀 Features

Secure Authentication – Sign in with Google, Apple, or Phone Number using Clerk
Fake News Detection – Submit a news article or summary to receive a credibility score and analysis
Gemini-Powered Verification – Gemini API compares submitted content with verified facts to determine authenticity
Credibility Scorecard – Visual feedback showing probability of news being fake or real with reasoning
Verified News Section – Browse fact-checked news articles with filters for politics, science, sports, and more
Search & Filter System – Easily locate news using keywords or interest-based categories
User Feedback System – Users can rate the accuracy of detections and submit feedback for improvement
Interactive Features – Enjoy quizzes and games related to fake news and misinformation
Community Section – Users can share news they find, explore community posts, like and interact with content
Modern UI/UX – Clean, intuitive, and responsive interface across all devices


🛠️ Tech Stack

  • Frontend: React.js, Next.js, Tailwind CSS
  • Backend: Flask, Node.js (for integration and routing)
  • Authentication: Clerk (Google, Apple, Phone Number login)
  • Database: MongoDB (user data, news history, community posts, feedback)
  • AI Integration: Gemini API (for content comparison and credibility scoring)
  • News & Fact-Check Sources: News API, Google Fact Check API
  • Deployment: Vercel (Frontend), Local backend (currently not cloud-hosted)

🧠 How Faktify Works

1️⃣ User Interaction Layer (Frontend)

  • User logs in securely via Clerk
  • Inputs news title, summary, or link
  • Receives a credibility score and breakdown
  • Can access verified news feed, play games, or explore community-shared content

2️⃣ Backend Processing

  • News input sent to Gemini API for evaluation
  • Gemini compares it against real-world data using Fact Check & News APIs
  • Returns a detailed credibility scorecard
  • Stores user interactions and feedback in MongoDB

3️⃣ Community & Feedback System

  • Logged-in users can submit suspicious news, which others can view, search, and like
  • Each result includes a comparison summary and community insights
  • Feedback data is stored and reviewed for future improvements

📸 Interface Previews (Snapshots)

🔗 Live Demo
(Includes detection flow, community page, and interactive quizzes)


🧪 Installation & Setup

Step 1: Clone the Repository

git clone https://github.com/yourusername/Faktify.git
cd Faktify

Step 2: Install Dependencies

# Frontend
npm install

# Backend
pip install -r requirements.txt

Step 3: Add Environment Variables

Create a .env file and add:

CLERK_API_KEY=your_clerk_api_key
MONGO_URI=your_mongodb_connection
NEWS_API_KEY=your_newsapi_key
FACTCHECK_API_KEY=your_google_factcheck_api_key
GEMINI_API_KEY=your_gemini_api_key

Step 4: Run the Application

# Frontend
npm run dev

# Backend
python app.py

👥 Team Members

Name Role GitHub LinkedIn
Vaibhav Gupta Team Lead, Full Stack Dev GitHub LinkedIn
Nishika Kansal Frontend Developer GitHub LinkedIn
Giriraj Mantri AI/ML & Integration Lead GitHub LinkedIn
Isha Katiyar Backend Developer GitHub LinkedIn

🛠️ Contribution Guidelines

We welcome open-source contributions!

  1. Fork the repo
  2. Create a new branch (feature/your-feature)
  3. Commit your changes
  4. Push and submit a Pull Request

📌 Roadmap & Upcoming Features

  • Multi-language support
  • Browser extension for quick fact-checking
  • User-based credibility scoring (trust scores)
  • Admin panel to monitor community posts & feedback

📃 License

This project is under the MIT License.


🌟 Star this project if you found it helpful — it motivates us to keep improving!

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •