Nepali Fake News Detector 🇳🇵

An AI-powered web application that detects whether a news article is REAL or FAKE, designed for Nepali and English content.

---

🌐 Live Demo

👉 https://demo.purwanews.com/

---

🚀 Features

• Fake / Real prediction
• Confidence score
• Nepali + English support
• Modern UI
• Real-time API

---

🧠 How it works

User Input → API → ML Model → Prediction

• TF-IDF Vectorizer
• Logistic Regression
• Returns prediction + confidence

---

📊 Dataset Used

The dataset used in this project is sourced from Kaggle:

👉 https://www.kaggle.com/ashoknepal/datasets

---

🧠 Dataset Description

• Collection of Nepali news articles
• Includes labeled data for fake and real news
• Contains fields such as:
  • news_context (news content)
  • label (0 = fake, 1 = real)

---

⚙️ Data Processing

To prepare the dataset for training:

• Multiple dataset files were combined
• Only required columns were used:
  • news_context
  • label
• Removed:
  • null values
  • empty text
  • duplicate entries
• Data was shuffled for better training

---

📌 Final Dataset Format

news_context	label
News text...	0
News text...	1

---

⚠️ Note

• Dataset quality directly affects model performance
• Results may vary on real-world news inputs

🛠 Tech Stack

• Python (Flask)
• Scikit-learn
• HTML, CSS, JavaScript
• cPanel Deployment

---

📂 Project Structure

fake-news-api/ │ ├── api.py ├── passenger_wsgi.py ├── train_model.py ├── model.pkl ├── vectorizer.pkl

---

📊 Model Accuracy

• Accuracy: ~96%
• Balanced performance on real and fake news

---

⚠️ Disclaimer

This system provides AI-based predictions and may not always be 100% accurate.

---

👨‍💻 Author

Santosh Poudel