Hard⏱ time3 hr$ cost0
Fake News Detector
Train a model to flag likely-misleading headlines from real ones.
Start building ↓
Result first
The build
Step 01
Clean headlines
Prepare and label the text.
Step 02
Vectorise
Turn words into features (TF-IDF).
Step 03
Train
Fit a classifier on real vs fake.
Step 04
Test
Check new headlines.
Working Principle
The model learns word and phrasing patterns more common in misleading text, then scores new headlines against them.
The science behind it
A closer look
It's text classification; results are only as good (and as biased) as the labelled data — a key lesson in responsible AI.
Take it further
Variables to test
- 1 Find a real headline it wrongly flags — why?
- 2 Discuss the risks of trusting it blindly.
More AI
AI
Hand-Drawn Digit Classifier
Train a tiny neural network in the browser to recognise digits you draw with a mouse.
Medium⏱2 hr$0
AI
Rock-Paper-Scissors Vision
Use a webcam and a no-code model to recognise your hand and play rock-paper-scissors.
Easy⏱1 hr$0
AI
Spam Email Classifier
Train a model on example messages so it can flag new ones as spam or not.
Medium⏱2 hr$0