Analyze · Grades 9-12
Confusion Matrix
Understand true positives, false negatives, and how we grade AI models.
The AI idea behind it
A confusion matrix is a small table that shows how a classifier's predictions compare with the truth. For a yes/no task it has four boxes: true positives, false positives, true negatives and false negatives.
It matters because accuracy alone can mislead. If only a few emails in a hundred are spam, a filter that never flags anything is still "mostly right" while missing every spam email. Precision and recall, both read from the matrix, show what accuracy hides.
How it works in the app
- See a model's predictions laid out as true positives, false positives, true negatives and false negatives.
- Read how the model is graded from the table, then see what a single accuracy score leaves out.
- Think through the cost of each kind of mistake for the task in front of you.
Try this
Which mistake is worse?
- Imagine two tasks: a spam filter and a test that screens for an illness.
- For each task, decide which is worse: a false positive or a false negative. Write down why.
- Explain how you would tune each model differently because of that answer.
What it shows: There is no single "best" model. The right trade-off depends on what each mistake costs the people affected.
Where it fits in the curriculum
More Analyze tools
Try Confusion Matrix in the app
One of the creation tools in LittleAIMaster for Grades 6-12.