Practice Questions: Modelling | Artificial Intelligence (AI) for Class 9 PDF Download

Q1. Which of the following is the broadest term encompassing all others?
(a) Machine Learning
(b) Deep Learning
(c) Artificial Intelligence
(d) Neural Networks

Ans: (c) 
AI is the umbrella term that includes both Machine Learning and Deep Learning.

Q2. What is the primary difference between Rule-Based and Learning-Based AI models?
(a) Rule-Based models self-learn, Learning-Based do not
(b) Rule-Based models follow static rules; Learning-Based models adapt to new data
(c) Learning-Based models follow developer rules
(d) Rule-Based models are always more accurate

Ans: (b) 
Rule-Based models use fixed rules set by developers, while Learning-Based models evolve by learning from new data.

Q3. What type of learning approach is used when a machine trains itself using large datasets and complex algorithms?
(a) Artificial Intelligence
(b) Machine Learning
(c) Rule-Based Modeling
(d) Deep Learning

Ans: (d)
Deep Learning is a subset of ML that uses large datasets to train itself without explicit programming.

Q4. What does 'Evaluation' in the AI project cycle refer to?
(a) Choosing datasets
(b) Writing algorithms
(c) Testing model performance with test data
(d) Implementing AI in apps

Ans: (c)
Evaluation checks the reliability and performance of a trained AI model using a separate testing dataset.

Q5. Which of the following is an example of a True Positive in a forest fire detection model?
(a) Fire predicted, no fire in reality
(b) Fire not predicted, fire in reality
(c) Fire predicted, fire in reality
(d) No fire predicted, fire in reality

Ans: (c)
This is a True Positive as the model correctly predicted a forest fire when there actually was one.

Q6. In which stage of the AI Project Cycle is a solution integrated into real-world use?
(a) Modeling
(b) Evaluation
(c) Data Exploration
(d) Deployment

Ans: (d)
Deployment is when the AI model is implemented in a real-world environment (e.g., chatbot on a website).

Q7. What is a major risk if the same dataset is used for both training and evaluation?
(a) Underfitting
(b) Deployment failure
(c) Ethical bias
(d) Overfitting

Ans: (d)
Using the training set for evaluation causes the model to simply memorize data, reducing real-world accuracy.

Q8. Which of the following is NOT a principle of AI ethics discussed in the module?
(a) Privacy
(b) Bias
(c) Accuracy
(d) Human Rights

Ans: (c)
The four key ethical principles discussed are Human Rights, Bias, Privacy, and Inclusion.

Q9. What ethical issue is highlighted when an AI grades an essay incorrectly due to biased training data?
(a) Privacy
(b) Inclusion
(c) Bias
(d) Deployment

Ans: (c)
If the training data is biased, the AI model may produce biased outcomes like unfair grading.

Q10. What is the best definition of 'Modelling' in AI?
(a) Creating rules for manual decision making
(b) Building and training AI algorithms
(c) Plotting graphs to visualize data
(d) Collecting raw datasets

Ans: (b)
Modeling refers to the process of creating and training AI models to learn from data.