Which is the best evaluation parameter considered in a loan approval model and why?

In loan approval, precision can be dangerous because it considers false positive rate when a person is not eligible for a loan but a machine predicts that the person is eligible.

The same problem is with recall which considers false negative rate because such evaluation can cause a loss to the bank because a person is eligible for a loan but the machine is predicting that he is not.

In such cases, F1-score can be a good evaluation technique because it maintains a balance between precision and recall and can tell almost exactly whether a person is eligible for loan or not.

The best evaluation technique is the accuracy which can tell us the best prediction because it gives the ratio of true prediction to total actual values.