Exam AWS Certified Machine Learning - Specialty topic 1 question 19 discussion

Ans. A is correct

Answer is A. An ROC curve (receiver operating characteristic curve) is a graph showing the performance of a classification model at all classification thresholds

Explanation: The ROC curve is the best technique to evaluate how different classification thresholds impact the model's performance. It plots True Positive Rate (TPR) against False Positive Rate (FPR) at various threshold values. Why the ROC Curve? Logistic regression outputs probabilities, and we need to select a classification threshold to decide between "order pizza" (1) and "not order pizza" (0). Changing the threshold impacts the trade-off between sensitivity (recall) and specificity. The ROC curve helps visualize this trade-off and select the best threshold based on the business goal (e.g., maximizing recall vs. minimizing false positives). The Area Under the ROC Curve (AUC-ROC) is a useful metric to measure the model's discrimination ability.

A is indeed correct see https://developers.google.com/machine-learning/crash-course/classification/roc-and-auc An ROC curve (receiver operating characteristic curve) is a graph showing the performance of a classification model at all classification thresholds. This curve plots two parameters: • True Positive Rate • False Positive Rate True Positive Rate (TPR) is a synonym for recall and is therefore defined as follows: TPR = TP/TP+FN False Positive Rate (FPR) is defined as follows: FPR = FP/FP+TN

The reason for this choice is that a ROC curve is a graphical plot that illustrates the performance of a binary classifier across different values of the classification threshold1. A ROC curve plots the true positive rate (TPR) or sensitivity against the false positive rate (FPR) or 1-specificity for various threshold values2. The TPR is the proportion of positive instances that are correctly classified, while the FPR is the proportion of negative instances that are incorrectly classified.

ROC curve is for defining the threshold.

A surely

Question is about classification so confusion matrix would come into mind; A is the answer

It is A.

obviously A

Root Mean Square Error (RMSE)　Ans. c