You have developed a logistic regression model to classify email spam and non$-$spam messages. The model achieves a high accuracy, but you

suspect that accuracy alone may not be a sufficient metric to evaluate the model's performance.

Explain why accuracy might be misleading in the context of email spam classification. Discuss additional performance metrics such as precision,

recall, F$1$ score, and ROC$-$AUC. How would you use these metrics to gain a more comprehensive understanding of your model's performance, and

which metrics would you prioritize for this specific task?