Logic Lens

A Chrome extension that detects logical fallacies and helps users think critically while reading online content

Comment

Logic Lens

Empowering Minds to Think Deeper.

Inspiration

In a world overwhelmed by information, arguments are everywhere — social media, news articles, debates, and everyday conversations. Yet many rely on flawed reasoning that often goes unnoticed.

We asked ourselves:

What if people could instantly recognize faulty logic while reading?

Logic Lens was built to inspire individuals to think deeper by making logical fallacies visible in real time.

What It Does

Logic Lens is a Chrome extension that analyzes highlighted text and detects logical fallacies.

When a user highlights a sentence, the system:

  1. Sends the text to a local Python server
  2. Runs it through a trained ML model
  3. Returns:
  • Fallacy type
  • Confidence score
  • Explanation

Currently supported fallacies include:

  • Ad Hominem
  • False Dilemma
  • Slippery Slope
  • Strawman
  • No Fallacy

How We Built It

Frontend

  • Chrome Extension (Manifest V3)
  • JavaScript, HTML, CSS
  • Content scripts for text capture
  • Background service worker

Backend

  • Python HTTP server (bridge.py)
  • JSON-based API communication
  • Local CSV storage

Challenges We Faced

  • Model version conflicts between scikit-learn versions
  • Feature size mismatches during retraining
  • Directory inconsistencies when loading serialized models
  • Frontend–backend communication debugging (CORS & localhost issues)

Many challenges came from integration and environment management rather than model design.

What We Learned

  • Model consistency across environments is critical
  • Feature engineering significantly impacts performance
  • Pickled models are fragile across library versions
  • Integration bugs often outweigh algorithmic complexity

Most importantly, we learned that AI can support critical thinking — not replace it.

What's Next for Logic Lens

  • Expand the training dataset
  • Add more fallacy categories
  • Improve confidence calibration
  • Explore transformer-based models
  • Deploy a cloud-hosted inference API

Final Thought

Logic Lens is more than a classifier — it’s a tool for intellectual growth.

Think Deeper. Question Better.

If you’d like, I can also:

  • Make this more technical (judge-focused)
  • Make it shorter (for skimmers)
  • Or make it more inspirational and mission-driven 🚀
Share this project:

Updates