Page 137 - Ai_V1.0_Class9
P. 137

TP vs FP rate at one    ROC Curve
                                        decision threshold
                          TP vs FP rate at                      The Receiver Operating  Characteristic (ROC) curve is  a
                          another decision                      graphical representation that illustrates the performance of a
                      1   threshold
                                                                binary classifier system at varying threshold values. It plots the
                                                                True Positive Rate (TPR) against the False Positive Rate (FPR) at
                                                                various threshold settings.

                       TP Rate                                  This curve plots two parameters:
                                                                     True Positive Rate (Sensitivity) is the proportion of actual
                                                                   •
                                                                   positive cases that are correctly identified by the classifier.
                                                                                              TP
                                                                                      TPR =
                                                                                            TP + FN
                      0                                            •    False  Positive  Rate is  the proportion of actual negative
                        0             FP Rate       1              cases that are incorrectly classified as positive.

                                                              FPR =    FP
                                                                    FP + TN
                 To generate an ROC curve, you need to perform the following tasks:
                    • Vary the threshold of your classifier, usually ranging from 0 to 1, and calculate TPR and FPR at each threshold.

                    • Plot these TPR and FPR values on a graph. TPR is plotted on the y-axis, and FPR is plotted on the x-axis.
                 Lowering the classification threshold classifies more items as positive, thus increasing both false positives and true
                 positives. The following figure shows a typical ROC curve.


                 Evaluate: Exoplanet Use Case

                 At this particular stage, we may need to
                 evaluate the model to find out the accuracy of
                 the model that makes the best prediction.

                 ROC  is  a  metric  that  is  used  to  find  out  the
                 accuracy of the model.
                 Figure shows the accuracy of the model using
                 ROC curve.


                         What is Deployment?


                 The deployment phase is the last stage of the AI project cycle and is used when the AI model is put into use in
                 a real-world setting. This involves integrating the model into existing systems or applications, such as creating
                 Application Programming Interfaces (APIs) or embedding it directly into software. It also includes setting up the
                 necessary infrastructure, like servers or cloud services, to support the model. Once integrated, the model needs
                 to be able to process new data and provide predictions. Monitoring tools are established to track the model’s
                 performance and ensure it works correctly. Logging and reporting are also important to capture data on how
                 the model is performing and to identify any issues that might arise. This phase is crucial for making the AI model
                 functional and useful for end-users.

                 Deployment of an  AI project  is an  essential phase  in bringing  the  created  AI solution into  practical  use.  The
                 following are some of the primary reasons why AI project deployment is relevant:
                    • Deployment converts theoretical models  into actual  tools that  are capable of being used in real-world
                   circumstances, bringing the AI solution into practice.
                                                                         AI Reflection, Project Cycle and Ethics  135
   132   133   134   135   136   137   138   139   140   141   142