Model Accuracy: How Good Is Your AI?
Model accuracy is the percentage of predictions the model gets correct. For example, if a model correctly classifies 87 out of 100 emails as spam or not spam, its accuracy is 87%.
About This Resource
Learn how to measure how well an AI model performs — accuracy percentages, true/false positives, and how to read a confusion matrix.
ML model evaluation resource suitable for Class 8–10 students.
What Students Will Learn
- Calculate accuracy as a percentage of correct predictions
- Define true positives, false positives, true negatives, and false negatives
- Read and interpret a simple confusion matrix
- Explain why high accuracy does not always mean a good model
Resources Available
Questions & Answers
What is model accuracy in machine learning?
Model accuracy is the percentage of predictions the model gets correct. For example, if a model correctly classifies 87 out of 100 emails as spam or not spam, its accuracy is 87%.
What is a confusion matrix?
A confusion matrix is a table that shows how many predictions were correct and how many were wrong, broken down by category. It reveals not just how many mistakes the model made, but what kind of mistakes.
What is a false positive?
A false positive is when the model predicts something is true, but it is actually false. For example, a spam filter that marks a legitimate email as spam has made a false positive.
Can a model have high accuracy but still be bad?
Yes! If 99% of emails are not spam, a model that labels everything 'not spam' would get 99% accuracy — but it would never catch a single spam email. This is why we use precision, recall, and F1-score alongside accuracy.
Related Resources
Browse resources by class: