What is the main limitation of using accuracy as an evaluation metric?
Practice Questions
Q1
What is the main limitation of using accuracy as an evaluation metric?
It does not account for false positives and false negatives
It is only applicable to regression problems
It requires a large dataset to be effective
It is difficult to calculate
Questions & Step-by-Step Solutions
What is the main limitation of using accuracy as an evaluation metric?
Step 1: Understand what accuracy means. Accuracy is the percentage of correct predictions made by a model out of all predictions.
Step 2: Recognize what class imbalance is. Class imbalance occurs when one class (or category) has many more examples than another class.
Step 3: Realize that in a class imbalance situation, a model can predict the majority class very well and still have high accuracy.
Step 4: Note that accuracy does not show how well the model performs on the minority class, which can be very important.
Step 5: Understand that because of this, accuracy can be misleading. A model might seem good because it has high accuracy, but it might be failing to predict the minority class correctly.
