Scenario: Red Lion National Bank, an offshore financial institution, is managed by Felix Manning and Wilson Fisk. The bank is migrating its operations to Microsoft Azure and has hired you as an Expert Consultant to oversee the IT team and provide guidance as needed. Currently, the IT team is developing a classification model to detect potential instances of credit card fraud. Project Details: The model is being built using Azure Machine Learning with automated machine learning (AutoML). The training dataset is highly unbalanced, meaning the number of fraudulent transactions is much smaller compared to legitimate ones. Felix has requested your expert advice on evaluating the classification model effectively. Question: Which primary metric should be used to evaluate the classification model's performance, given the imbalanced dataset?