TruthPulse

A Flask web app that detects fake news using a TF-IDF + Passive Aggressive Classifier model, backed by PostgreSQL for logging and history.

Features

• Fake news classification via TF-IDF vectorization and a Passive Aggressive Classifier
• PostgreSQL-backed history of past checks
• Simple web UI (submit text, get a result, view history)

Tech Stack

• Python, Flask
• scikit-learn, pandas, numpy, joblib
• PostgreSQL (psycopg2)

Setup

git clone https://github.com/zulqarnain2004/TruthPulse.git
cd TruthPulse/"dbms project"
pip install -r requirements.txt

Create a PostgreSQL database named actchecker, then copy .env.example to .env and fill in your own credentials:

copy .env.example .env

Run the app:

python app.py

Project Structure

dbms project/ ├── app.py # Flask routes ├── ai_model.py # TF-IDF + PassiveAggressiveClassifier model ├── database.py # PostgreSQL connection and logging ├── templates/ # HTML views ├── static/ # CSS └── requirements.txt

License

MIT