Page 236 - AI Ver 3.0 class 10_Flipbook
P. 236

Total no. of correct predictions
                                     Classification Accuracy =                             × 100
                                                                 Total no. of predictions
                                                                 (TP + TN)
                                                          =                    × 100
                                                             (TP + TN + FP + FN)
                                                             25
                                                          =     × 100
                                                             51
                                                          = 49%

              Can we use Accuracy all the time?

              It is suitable wherever the dataset is balanced, which means the positive and negative classes are roughly equal, that
              is a rare occurrence, and that all predictions and prediction errors are equally important, which is often not the case.
              For example, Calculating the accuracy of the classifier model, that predicts whether a student will pass a test
              (Yes) or not pass a test (No). It classifies the input into two classes Yes and No. Let's, calculate the accuracy of the
              classifier model and construct the confusion matrix for the model.
              Here,
                 • Total test data is 1000.

                 • Actual values are 900 Yes and 100 No (Unbalanced dataset).
                 • It is a faulty model which, irrespective of any input, will give a prediction as Yes.
                 • Calculate the classification accuracy of this model.
              To prepare the classification accuracy of this model follow the given steps:

               Step 1    Construct the Actual value vs Predicted value table. Consider Yes as the positive class and No as the
                        negative class.


                                                Predicted Value       Actual Value








               Step 2   Construct the confusion matrix.
                          So, the faulty model will predict all the 1000 input data as Yes.
                           Consider Yes as the positive class and No as the negative class. Construct the confusion matrix from the
                       Actual vs Predicted table.

                                                  Predicted Value    Actual value

                                                     Yes=1000          Yes=900
                                                       No=0             No=100

                                                              Predicted Values
                                                             Yes         No
                                                      Actual Values  Yes  No  TP=  FN=




                                                             FP=
                                                                        TN=

                    234     Touchpad Artificial Intelligence (Ver. 3.0)-X
   231   232   233   234   235   236   237   238   239   240   241