Databricks Generative AI Engineer Associate Question 171
Single answerYou are building a text classification system on Databricks and have fine-tuned several generative AI models. After running experiments, you have the following evaluation metrics for the models. Which model would you select if the primary business goal is to minimize false positives while maintaining a reasonably high precision?
Model A: Precision = 0.85, Recall = 0.65 Model B: Precision = 0.92, Recall = 0.50 Model C: Precision = 0.78, Recall = 0.85 Model D: Precision = 0.80, Recall = 0.60
- A
Model A
- B
Model B
- C
Model C
- D
Model D
Show answer and explanation
Correct answer: B
Explanation
Precision is the key metric for minimizing false positives, as it measures the proportion of true positive predictions out of all positive predictions. Among the models, Model B has the highest precision (0.92), making it the best choice for this specific business goal, even though its recall is relatively low. In contrast, other models have lower precision, making them less effective in minimizing false positives.
- A. Incorrect.
Model A has a precision of 0.85, which is good, but it does not minimize false positives as effectively as Model B, which has a higher precision.
- B. Correct.
Model B has the highest precision (0.92), which directly correlates to minimizing false positives. Although its recall is lower, the primary business goal is to minimize false positives rather than maximize recall.
- C. Incorrect.
Model C has a high recall (0.85), which is beneficial for minimizing false negatives, but its precision is lower (0.78), making it less suitable for minimizing false positives.
- D. Incorrect.
Model D has a precision of 0.80, which is lower than Model B, and therefore does not minimize false positives as effectively.