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
   226   227   228   229   230   231   232   233   234   235   236