Page 231 - AI Ver 3.0 class 10_Flipbook
P. 231
• True Negative: The predicted value matches the actual value i.e.; the actual value was negative and the model
predicted a negative value.
• False Positive (Type 1 error): The predicted value was falsely predicted i.e.; the actual value was negative but
the model predicted a positive value.
• False Negative (Type 2 error): The predicted value was falsely predicted i.e.; the actual value was positive but
the model predicted a negative value.
Brainy Fact
AI algorithms can analyse weather patterns and other data to predict natural disasters such as hurricanes and
earthquakes.
For example, Loan Approval
Let’s say, based on some metric parameters; you have designed a classifier that predicts whether a loan will be
approved or not.
The output is 1 if the loan is approved or 0 if loan is not approved or rejected. That is, 1 and 0 signify whether there
loan is approved or not.
The following is a confusion matrix of models predicting whether the loan is approved or not.
Predictions for 1 that were actually 1 Predictions for 0 that were actually 1
appear in the cell. Implying prediction Predicted Values appear in the cell. Implying prediction
that the loan will be approved and the 1 0 that the loan will not be approved
loan was approved actually. and the loan was approved actually.
Actual Values 1 20 10
Predictions for 1 that were actually 0 0 15 5 Predictions for 0 that were actually 0
appear in the cell. Implying prediction appear in the cell. Implying prediction
that the loan will be approved and the that the loan will not be approved and
loan was not approved actually. the loan was not approved actually.
Predicted
Confusion Matrix
Yes No
Yes 20 10
Actual
No 15 5
Brainy Fact
Mammography is the most effective way to detect breast cancer, but it can miss it 20% of the time. AI can help
improve the accuracy of breast cancer detection and prediction. An AI model developed by researchers at MIT’s
Computer Science and Artificial Intelligence Laboratory (CSAIL) and Jameel Clinic for Machine Learning. Mirai is
a complex neural network that can detect breast cancer up to five years before it occurs.
Evaluating Models 229

