Teach an AI with ExamplesSupervised Learning & Data Quality
Discover how AI learns from labeled examples, experiment with balanced vs unbalanced data, and train your own simulated AI classifier!
How AI Learns
Just like a child learns to recognize animals from picture books, an AI learns from thousands of labeled training examples.
Training Example
A single input data point shown to the AI (e.g. one email message).
Target Label
The ground truth answer paired with the example (e.g. 'SPAM').
Training Set
The complete collection of labeled example pairs used to train the model.
Analogy: If you show a student 500 cat photos labeled 'CAT' and 500 dog photos labeled 'DOG', they quickly learn the difference. AI algorithms learn statistical patterns from datasets in the exact same way!
Looking at Labeled Data
Here are examples of labeled training data for an email spam filter. Notice the patterns:
The AI spots pattern signals: repeated exclamation points (!!!), ALL-CAPS words (FREE, WIN, NOW), and urgent calls to action.
Data Quality Matters
More data usually makes a smarter AI, but the quality and balance of data matter just as much!
More is Better
10,000 examples are far better than 10 examples. AI needs enough data to find real patterns.
Balanced is Better
Equal SPAM and NOT SPAM examples prevent the AI from becoming biased toward one answer.
Diverse is Better
Cover different types of messages so the AI generalizes well to unseen real-world data.
Playground: Train Your Spam Filter
Experience DATA → TRAIN → MODEL → TEST → PREDICTION. Add your own training examples, try the Data Quality experiment, and test the simulated AI!
💡 Note: This is a beginner-friendly simulation of ML pattern recognition.
Training Dataset (4 Total Examples)
Test the AI (Predict New Message)
💡 Explanation: Detected high-risk spam keywords: 'won', 'free', 'gift'.
Interactive Exercises
Exercise 1: Label These Examples
+15 XP · BeginnerExercise 2: Good Training Data
+20 XP · BeginnerExercise 3: Data Quality Problem
+25 XP · IntermediateExercise 4: Too Little Data
+30 XP · IntermediateKnowledge Check
Ready to test your knowledge?
Answer 10 multiple-choice questions to test your understanding of labeled data and AI training!
Build Real Machine Learning Models with Tutors
Take the next step! Join our live Intro to Machine Learning Course and write real Python ML code with expert 1-on-1 guidance.