Page 152 - Ai_C10_Flipbook
P. 152

False Positive
                                                                                                        Predicted
              A  False  Positive  occurs  when  a  model  incorrectly  predicts  a   Confusion Matrix
              positive  outcome  for  a  case  that  is  actually  negative.  When  a                 Yes       No
              model’s prediction does not match with the actual outcome. In the             Yes        3         2
              given example the shaded region depicts the False Positive.        Actual
                                                                                            No         2         3
              Some more examples of False Positive are as follows:
                 • Hiring Systems (AI-based Recruitment) - An AI system screens suitable job applicants.
                False Positive is a less-qualified candidate is identified as a good match.
                 • Autonomous Vehicles (Object Detection) - A self-driving car detects obstacles on the road.

                False Positive is the car incorrectly identifies a harmless shadow as an obstacle.
              False Negative


              A False Negative occurs when a model incorrectly predicts a negative outcome when the true outcome is actually
              positive. When a model’s prediction does not match with the actual outcome. In the given example the shaded
              region depicts the False Negative.
              Some more examples of False Negative are as follows:
                                                                                                        Predicted
                 • Security Systems - A facial recognition system does not identify   Confusion Matrix
                 family members as intruders.                                                         Yes       No
                False Negative is an intruder is recognised as a family member.             Yes        3         2
                                                                                Actual
                 • Fire  Alarm  Systems  -  On  any  normal  day,  a  fire  detection       No         2         3
                 system will not trigger fire alarm.
                False Negative: The system detects a fire to detect a fire in case of no fire.

              The confusion matrix with all classification outcomes based on the different values of actual and predicted labels
              can be presented as follows:


                                                                        Predicted
                                       Confusion Matrix
                                                                 YES                NO
                                                             True Positive     False Negative
                                                   YES      Predicted – Yes    Predicted – No
                                                              Actual - Yes      Actual - Yes
                                       Actual
                                                             False Positive    True Negative
                                                   NO       Predicted – Yes    Predicted – No
                                                              Actual - No       Actual - No


                       Accuracy from Confusion Matrix


              Classification Accuracy is the percentage of correct predictions out of the total observations made by an AI
              model. It provides a clear picture of how accurate the predictions are for the given model. A high accuracy score
              generally indicates good performance, as it accounts for all correctly predicted values. The mathematical formula
              for classification accuracy is:


                                                               No. of correct predictions
                                      Classification Accuracy =                           × 100
                                                                Total no. of predictions

                    150     Artificial Intelligence Play (Ver 1.0)-X
   147   148   149   150   151   152   153   154   155   156   157