MLS-C01 Question 232
Single answerYou are building a binary classification model to predict whether users will subscribe to a service (1: Yes, 0: No). After evaluating the model, you find the following confusion matrix:
| Predicted: No | Predicted: Yes | |
|---|---|---|
| Actual: No | 950 | 50 |
| Actual: Yes | 100 | 900 |
What conclusion can you draw about the model's performance based on this confusion matrix?
- A
The model has a high precision for predicting users who will subscribe to the service.
- B
The model is underperforming because it misclassifies a large number of actual 'No' cases as 'Yes'.
- C
The model has a high recall for predicting users who will not subscribe to the service.
- D
The model performs poorly overall because its accuracy is below 50%.
Show answer and explanation
Correct answer: A
Explanation
The confusion matrix shows that the model has a high precision for predicting 'Yes' cases (subscribers). Precision focuses on the proportion of correctly predicted 'Yes' cases out of all cases predicted as 'Yes', and in this case, it is 94.7%. The other options either misinterpret the confusion matrix or focus on metrics not relevant to the question.
- A. Correct.
Correct. Precision for predicting 'Yes' is calculated as True Positives / (True Positives + False Positives). Here, it is 900 / (900 + 50) = 0.947 or 94.7%, which is high.
- B. Incorrect.
Incorrect. The model misclassifies 50 'No' cases as 'Yes', but this is a small portion compared to the 950 correctly classified 'No' cases. The model does not have a significant issue with false positives.
- C. Incorrect.
Incorrect. Recall for predicting 'No' is calculated as True Negatives / (True Negatives + False Negatives). Here, it is 950 / (950 + 100) = 0.905 or 90.5%, which is high but not related to this question.
- D. Incorrect.
Incorrect. The model's overall accuracy is (950 + 900) / (950 + 50 + 100 + 900) = 0.925 or 92.5%, which is well above 50%.