https://github.com/prajwalsde/fake-news-detection

A Machine Learning-based Fake News Detection system using TF-IDF and Logistic Regression, with a Streamlit app for real-time predictions.
https://github.com/prajwalsde/fake-news-detection

data-science fake-news-detection machine-learning news-classifier nlp python sklearn streamlit text-classification tfidf

Last synced: 4 months ago
JSON representation

A Machine Learning-based Fake News Detection system using TF-IDF and Logistic Regression, with a Streamlit app for real-time predictions.

• Host: GitHub
• URL: https://github.com/prajwalsde/fake-news-detection
• Owner: prajwalsde
• License: mit
• Created: 2025-06-22T13:12:32.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2025-06-22T13:44:35.000Z (4 months ago)
• Last Synced: 2025-06-22T14:33:10.783Z (4 months ago)
• Topics: data-science, fake-news-detection, machine-learning, news-classifier, nlp, python, sklearn, streamlit, text-classification, tfidf
• Language: Python
• Homepage: https://fake-news-detection-sbaybxhb8jtijbcfw9zmol.streamlit.app
• Size: 41.5 MB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Fake News Detection using Machine Learning

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

![Python](https://img.shields.io/badge/Python-3.10%2B-blue)

![Streamlit](https://img.shields.io/badge/Built%20with-Streamlit-orange)

![Platform](https://img.shields.io/badge/Platform-MacOS%20%7C%20Linux%20%7C%20Windows-informational)

![GitHub Repo stars](https://img.shields.io/github/stars/prajwalsde/fake-news-detection?style=social)

![GitHub forks](https://img.shields.io/github/forks/prajwalsde/fake-news-detection?style=social)

Detect whether a news article is fake or real using NLP and ML.

## Tech Stack

- Python, Pandas, Scikit-learn, Streamlit

- Logistic Regression with TF-IDF

- Dataset: Kaggle Fake and Real News

## 📌 Features

- ✅ Classifies news as **Real** or **Fake**

- 💬 Built using **TF-IDF Vectorizer** and **Logistic Regression**

- ⚡ Interactive **Streamlit app**

- 🧠 Trained on real-world datasets (`Fake.csv`, `True.csv`)

- 💾 Model saved with `pickle` for fast deployment  

## How to Run

1. Clone the repo

2. Install dependencies: `pip install -r requirements.txt`

3. Train model: run `notebook/main.ipynb`

4. Run app: `streamlit run app/app.py`

## Demo

Enter a news article in the text area, and the app will tell you if it's fake or real!