AI & ML WorksheetsModel Accuracy & Predictions

Free Interactive Worksheet · Beginner

Model Accuracy & PredictionsConfusion Matrix & Evaluation

Learn how to calculate AI model accuracy, build confusion matrices (TP, TN, FP, FN), and evaluate predictions interactively!

🎯

Overview

How do we measure AI success?

After training a Machine Learning model, we measure its real-world performance. The simplest and most common evaluation metric is Accuracy — the percentage of predictions the model got right out of total attempts.

The Accuracy Formula

Accuracy = (Correct Predictions / Total Predictions) × 100%

🧮 Interactive Accuracy Calculator

Adjust the total and correct predictions below to see how accuracy updates live:

Total Predictions: 20

Correct Predictions: 18

Calculated Model Accuracy:18 / 20 = 90%

📊

Accuracy Explained

Spam Filter Case Study

Imagine an AI spam filter tested on 20 new emails. It correctly flags 18 emails and makes 2 errors.

Notice: Accuracy measures overall correctness, but doesn't tell us WHICH types of errors were made (e.g. did it miss spam, or delete your boss's important email?).

❓ Quick Practice: If a model correctly predicts 18 out of 20 emails, what is its accuracy?

🧩

Confusion Matrix

Breaking Down the 4 Outcomes

To see exactly where predictions go right or wrong, we build a Confusion Matrix with 4 quadrants:

True Positive (TP) ✅

Spam email correctly identified and moved to Spam folder.

True Negative (TN) ✅

Real email correctly sent to your Inbox.

False Positive (FP) ❌ (False Alarm)

Real important email falsely marked as Spam!

False Negative (FN) ❌ (Missed Spam)

Spam email falsely predicted as safe, slipping into Inbox.

	Predicted: SPAM	Predicted: NOT SPAM
Actual: SPAM	True Positive (TP) Spam caught!	False Negative (FN) Spam missed!
Actual: NOT SPAM	False Positive (FP) Real email blocked!	True Negative (TN) Real email safe!

🎮

Playground: Confusion Matrix Builder

Classify & Track Live Predictions

Classify each email below as Spam or Not Spam to see the 2x2 confusion matrix and accuracy update live!

"Win a free $1000 gift card now!"

Actual: Spam

"Team meeting tomorrow at 3 PM"

Actual: Not Spam

"URGENT: Your bank account is suspended!"

Actual: Spam

"Dinner with family this weekend?"

Actual: Not Spam

"Click here for cheap medicine deals"

Actual: Spam

"Invoice for your recent book order"

Actual: Not Spam

"CONGRATULATIONS you have been selected!"

Actual: Spam

"Project status report update"

Actual: Not Spam

"Double your savings fast"

Actual: Spam

"Can you review this code snippet?"

Actual: Not Spam

Live Confusion Matrix

True Positive (TP)0

False Negative (FN)0

False Positive (FP)0

True Negative (TN)0

CURRENT ACCURACY0%0 / 0 Correct (0/10 Classified)

Round 1 Score: 0%Round 2 Score: 0%

💪

Interactive Exercises

Total XP Earned: 0 / 90 XP

0 / 4 Completed

Exercise 1: Calculate Accuracy

+15 XP · Beginner

Exercise 2: Identify Error Type 1

+20 XP · Intermediate

Exercise 3: Identify Error Type 2

+25 XP · Intermediate

Exercise 4: The Accuracy Illusion

+30 XP · Intermediate

🧠

Knowledge Check

Ready to test your knowledge?

Answer 10 multiple-choice questions to test your understanding of model accuracy and confusion matrices!

🚀 Ready for live coding?

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.

Explore ML Course →