Page 233 - AI Ver 3.0 class 10_Flipbook
P. 233
So, the final Confusion matrix will be as follows:
Predicted
Confusion Matrix
Yes No
Yes 3 2
Actual
No 2 3
Here, the total number of correct predictions are 6 out of 10
The classification outcomes based on the different values of actual and predicted labels are as follows:
• True Positive
• True Negative
• False Positive
• False Negative
True Positive
A True Positive occurs when a model correctly predicts a positive outcome. In the above example, the value of the
True Positive is depicted in the shaded region.
Predicted
Confusion Matrix
Yes No
Yes 3 2
Actual
No 2 3
Some more examples of True Positive are as follows:
• Medical Diagnosis - A machine learning model predicts whether the patient has asthma.
True Positive: The model predicts that the patient actually has asthma.
• Face Recognition Security System - A security system identifies individuals who are authorised to enter a
restricted area.
True Positive: The system recognises an authorized employee correctly.
True Negative
A True Negative occurs when a model correctly predicts a negative outcome. When the model’s negative
prediction is same as the actual outcome, it’s the case of True Negative. In the given example the shaded region
depicts the True Negative scenario.
Predicted
Confusion Matrix
Yes No
Yes 3 2
Actual
No 2 3
Evaluating Models 231

