Page 314 - Ai Book - 10

3. Under what circumstances might Accuracy alone be insufficient for evaluating the performance of an AI
model?

4. In the context of Scenario 1, predicting water shortage in schools, discuss the significance of a True
Negative.
5. How does the concept of Recall contribute to understanding the effectiveness of an AI model, and in
what types of applications is it particularly crucial?

G. Application based questions.
1. If you were to develop an AI model for predicting unexpected rain, explain how False Positive might
impact the model’s performance. Provide a practical scenario to illustrate the consequences of False
Positive in this application.
2. Apply the concept of model evaluation to an AI model designed to predict water shortages in schools.
Discuss how Precision, Recall, and F1 Score would contribute to understanding the model’s performance,
and why these metrics are relevant in the educational context.

3. Prepare a plan to deploy AI model for disaster management, specifically predicting floods. How is the
concept of a confusion matrix be applied in evaluating the model’s performance? Provide a step-by-step
explanation using terms like True Positive, True Negative, False Positive, and False Negative.
H. Assertion and reason-based questions.

1. Assertion: True Negative are crucial in model evaluation, especially in disaster prediction scenarios.
Reason: True Negative indicate instances where the model correctly predicts a negative reality.

a. True for both Assertion and Reason b. True for Assertion, False for Reason
c. False for Assertion, True for Reason d. False for both Assertion and Reason
2. Assertion: A confusion matrix summarizes the distribution of prediction outcomes in a model.

Reason: It provides insights into True Positive, True Negative, False Positive, and False Negative.
a. True for both Assertion and Reason b. True for Assertion, False for Reason

c. False for Assertion, True for Reason d. False for both Assertion and Reason
3. Assertion: Using a test dataset that has never been part of training is essential for model evaluation.
Reason: It ensures the model’s ability to generalize to new data and prevents overfitting.

a. True for both Assertion and Reason b. True for Assertion, False for Reason
c. False for Assertion, True for Reason d. False for both Assertion and Reason

AI Fun Zone
21st
21st
Century
Century Project-based Learning
Skills
Skills
Prepare a project report on the topic ‘Predicting AI Models’.

188
188