Page 226 - AI Computer 10

G. Competency-based questions.

1. You have been assigned a task to deploy an AI-based fl ood predicti on model in a region prone to fl ooding.
Which metric would be most criti cal in evaluati ng the performance of the model?

a. Accuracy b. Recall
c. Precision d. F1 Score

2. Consider deploying an AI model for medical diagnoses of a disease which is not criti cal for a pati ent in
the long run, but its treatment includes medicati on that has severe side eff ects. Which of the following
metrics would best evaluate the model?

a. Accuracy b. Recall
c. Precision d. F1 Score

3. You are an analyst hired to provide technical and tacti cal support to a team in the IPL. The owners want
you to prepare an AI model to predict how the team will fare against an opposing team in a match. What
could be the training dataset for the model made of?

a. Records of all past matches of IPL in previous editi ons
b. Records of all past matches of the team in previous editi ons

c. Records of all cricket matches ti ll date
d. Records of all players in the team
Answers

1. (b) 2. (c) 3. (b)
H. Assertion and Reason-based questions.
In the following questi ons, a statement of Asserti on (A) is followed by a statement of Reason (R). Study both
the statements and give the answer as:

(a) if both Asserti on and Reason are correct and Reason is the correct explanati on of Asserti on.
(b) if both Asserti on and Reason are correct but Reason is not the correct explanati on of Asserti on.
(c) if Asserti on is True but Reason is False.

(d) if Asserti on is False but Reason is True.
1. Asserti on (A): Precision is an important metric in evaluati ng AI models.
Reason (R): Precision measures the accuracy of positi ve predicti ons, ensuring a low rate of false positi ves.
2. Asserti on (A): Accuracy alone may not be suffi cient for evaluati ng a model’s performance.

Reason (R): Accuracy does not consider the real-world scenarios where wrong predicti ons can be criti cal.
3. Asserti on (A): F1 Score is considered a bett er metric for model evaluati on than just acccuracy.

Reason (R): F1 Score takes into account both Precision and Recall, preventi ng the dominance of one
parameter over the other.

Answers
1. (c) 2. (a) 3. (a)
I. Case-study based questions.

1. A company developed an AI model to classify emails as spam or not spam. The model was evaluated on
1000 test cases. The given confusion matrix summarises the predictions made by the model as compared
to the actual results.

92
92