Skip to content
Sangam Gupta
All work

NewsIntel

Paste a headline or article and get a real-or-fake verdict in seconds.

Status
Live
Stack
Python, scikit-learn, TF-IDF, NLTK, Flask, Supabase, Google OAuth
Links
Open NewsIntelCode

How it fits together

  1. News text
  2. NLTK cleaning
  3. TF-IDF vectoriser
  4. Passive Aggressive classifier
  5. Flask app
    • Real or fake verdict
    • Supabase history

My final-year engineering project: an NLP platform that classifies news as real or fake with TF-IDF features, NLTK preprocessing and a Passive Aggressive classifier, served through Flask with Google sign-in and verification history on Supabase.

The problem

Misinformation spreads faster than people can fact-check it. I wanted a tool that gives an instant first signal on whether a piece of news looks credible.

How I built it

  1. Clean labelled news text with NLTK and vectorise it with TF-IDF.
  2. Train a scikit-learn Passive Aggressive classifier on those features.
  3. Serve predictions through a Flask app, built with AI assistance, with Google OAuth 2.0 sign-in and each user's verification history in Supabase (PostgreSQL).
  4. Deploy on its own subdomain with a responsive interface.

Where it is now

Live at newsintel.sangamgupta.in.

Ask about NewsIntel

Answers come from this case study and its code on GitHub.