AN

MODEL TEST PAPER 3

2
(Based on Latest CBSE Sample Paper)

H
Max. Time: 2 Hours Max. Marks: 50

AS
General Instructions:
1. Please read the instructions carefully.
2. This Question Paper consists of 21 questions in two sections: Section A & Section B.

AK
3. Section A has Objective type questions whereas Section B contains Subjective type questions.
4. Out of the given (5 + 16 =) 21 questions, a candidate has to answer (5 + 10 =) 15 questions in the
allotted (maximum) time of 2 hours.

PR
5. All questions of a particular section must be attempted in the correct order.
6. SECTION A – OBJECTIVE TYPE QUESTIONS (24 MARKS):
(i) This section has 05 questions.
(ii) Marks allotted are mentioned against each question/part.
(iii) There is no negative marking.
(iv) Do as per the instructions given. S
7. SECTION B – SUBJECTIVE TYPE QUESTIONS (26 MARKS):
ER
(i) This section has 16 questions.
(ii) A candidate has to do 10 questions.
(iii) Do as per the instructions given.
H

(iv) Marks allotted are mentioned against each question/part.

SECTION A: OBJECTIVE TYPE QUESTIONS

OT

1. Answer any 4 out of the given 6 questions on Employability Skills (1 × 4 = 4 marks)

(i) “Vikram, an athlete, encountered an injury during a major tournament. Instead of getting
discouraged, he focused on his rehabilitation, stayed positive, and worked hard to recover.
BR

A few months later, he returned stronger and performed better than before.”
Which self-management skill is evident from Vikram’s approach?
Ans. The self-management skill evident from Vikram’s approach is "Resilience". His ability to stay
optimistic, bounce back from the injury, and improve his performance highlights his resilience.
(ii) Which of the following is the correct method to shut down a computer safely?
L

(a) Press and hold the power button

YA

(b) Unplug the power cable

(c) Use the "Shut Down" option from the Start menu
(d) Close all windows and leave the computer on
GO

Ans. (c) Use the "Shut Down" option from the Start menu
(iii) Assertion (A): Government employees often receive a pension after retirement.
Reason (R): A pension is a feature of wage employment.

Artificial Intelligence Class-10 (Subject Code-417) 1

 AN
(a) Both A and R are correct and R is the correct explanation of A
(b) Both A and R are correct but R is NOT the correct explanation of A
(c) A is correct but R is not correct
(d) A is not correct but R is correct

H
Ans. (a) Both A and R are correct and R is the correct explanation of A
(iv) The process of deciding what tasks are most important and doing them first is called:

AS
(a) Delegating (b) Prioritising
(c) Monitoring (d) Scheduling
Ans. (b) Prioritising
(v) John had difficulty collaborating with his team because he felt they didn’t respect his ideas

AK
due to his junior position. This is an example of
(a) Interpersonal barrier (b) Physical barrier
(c) Organisational barrier (d) Cultural barrier
Ans. (a) Interpersonal barrier

PR
(vi) “Efforts are being made to ensure everyone has access to safe and clean drinking water,
especially in rural and underserved areas.”
Which SDG does this relate to?
(a) Clean water and sanitation (b) Life below water
(c) Affordable and clean energy (d) Decent work and economic growth
Ans. S
(a) Clean water and sanitation
2. Answer any 5 out of the given 6 questions  (1 × 5 = 5 marks)
ER
(i) Assertion (A): A person can be strong in musical intelligence but not necessarily good at
logical reasoning.
Reason (R): Each individual has strengths and weaknesses based on their dominant types of
intelligence.
H

(a) Both A and R are correct and R is the correct explanation of A

(b) Both A and R are correct but R is not the correct explanation of A
OT

(c) A is correct but R is not correct

(d) A is not correct but R is correct.
Ans. (a) Both A and R are correct and R is the correct explanation of A
(ii) A tech company was criticized for its algorithms that unfairly favored certain demographics
BR

over others in its job recruitment tool.

Which terminology best describes this issue?
(a) AI Ethics (b) Data Privacy
(c) AI Access (d) AI Bias
Ans. (d) AI Bias
(iii) Statement 1: An activation function in a neural network introduces non-linearity into the
L

model.
YA

Statement 2: The loss function is used to measure the accuracy of the predictions made by
the neural network.
(a) Both Statement 1 and Statement 2 are correct
(b) Both Statement 1 and Statement 2 are incorrect
GO

(c) Statement 1 is correct but Statement 2 is incorrect

(d) Statement 2 is correct but Statement 1 is incorrect
Ans. (c) Statement 1 is correct but Statement 2 is incorrect

2 Model Test Paper 3 (Solved)

 AN
(iv) Observe the given graph and fill in the blank:
Hidden layer Hidden layer

H
AS
Input

ReLU ReLU

AK
PR
The layer that takes input data and applies weights and biases to generate output is known
as the ________________ layer.
Ans. Input
S
ER
(v) ________________ file format is primarily used for storing spreadsheet data, including cells,
formulas, and charts?
(a) csv (b) html
(c) xlsx (d) mp3
H

Ans. (c) xlsx

(vi) In a text analysis, if a word appears frequently across many documents, it is classified as:
(a) Rare word (b) Common word
OT

(c) Frequent word (d) Irrelevant word

Ans. (c) Frequent word
3. Answer any 5 out of the given 6 questions (1 × 5 = 5 marks)
BR

Read the examples given below:

1. Using a virtual assistant to set reminders
2. Automated email filters sorting messages
3. Temperature control in a room using a thermostat
4. Facial recognition for security access
L

(i) Choose the options that are not AI:

(a) (i) and (ii) (b) (iii) and (iv)
YA

(c) (ii) and (iii) (d) (i) and (iv)

Ans. (b) (iii) and (iv)
(ii) ________________ refer to the different attributes or characteristics that we use to describe and
analyze data.
GO

(a) Labels (b) Features

(c) Targets (d) Instances
Ans. (b) Features

Artificial Intelligence Class-10 (Subject Code-417) 3

 AN
(iii) Which of the following is a common use of data science in healthcare?
(a) Disease prediction (b) Social media marketing
(c) Virtual reality (d) Music streaming
Ans. (a) Disease prediction

H
(iv) ________________ refers to the technique used to recognize and categorize individual instances
of objects within an image.

AS
(a) Classification (b) Object detection
(c) Instance segmentation (d) Feature extraction
Ans. (c) Instance segmentation
(v) Identify the given Chatbot type: It uses predefined responses and follows a script to handle

AK
user interactions. This type typically does not learn from conversations or adapt based on
new information.
(a) Rule-based chatbot (b) AI-driven chatbot
(c) Hybrid chatbot (d) Conversational chatbot

PR
Ans. (a) Rule-based chatbot
(vi) The ROC curve is used to evaluate the performance of a classification model by plotting the
trade-off between
(a) True Positives and False Negatives (b) True Positives and False Positives
(c) Precision and Recall (d) False Positives and False Negatives
S
Ans. (b) True Positives and False Positives
ER
4. Answer any 5 out of the given 6 questions (1 × 5 = 5 marks)
(i) Ankit was studying AI concepts but forgot the specific term used for a type of artificial
intelligence that involves algorithms designed to simulate human neural processes. What is
this term?
H

(a) Machine Learning (b) Natural Language Processing

(c) Deep Learning (d) Reinforcement Learning
OT

Ans. (c) Deep Learning

(ii) Statement 1: Overfitting occurs when a model performs well on training data but poorly on
unseen data.
Statement 2: Underfitting happens when the model performs well on both training and test
data.
BR

(a) Both Statement 1 and Statement 2 are correct

(b) Both Statement 1 and Statement 2 are incorrect
(c) Statement 1 is correct but Statement 2 is incorrect
(d) Statement 2 is correct but Statement 1 is incorrect
Ans. (c) Statement 1 is correct but Statement 2 is incorrect
L

(iii) Ravi developed a predictive model to forecast sales for his company. He used historical sales
YA

data for training and a different set of recent sales data to test the model’s accuracy. Which
statement is incorrect?
(a) The historical sales data used for training the model is known as Training Data.
(b) The recent sales data used to evaluate the model’s performance is known as Testing Data.
GO

(c) Training Data and Testing Data are both part of the Data Preparation stage.
(d) Testing Data is used during the model training process.
Ans. (d) Testing Data is used during the model training process.

4 Model Test Paper 3 (Solved)

 AN
(iv) What is the typical number of layers in a basic feed forward neural network?
(a) 1 (b) 2
(c) 3 (d) 4

H
Ans. (c) 3 (Input layer, hidden layer(s), output layer)
(v) Which feature of NLP allows for the recognition and extraction of specific entities such as

AS
names and dates from a text?
(a) Automatic Summarization (b) Named Entity Recognition
(c) Sentiment Analysis (d) Text Classification
Ans. (b) Named Entity Recognition

AK
(vi) Priya created a spam filter model by training and testing it on the same set of emails. While
the model achieved excellent results with this data, it struggled with new, unseen emails.
What concept does this situation highlight?
(a) Overfitting (b) Data Augmentation
(c) Hyperparameter Tuning (d) Feature Engineering

PR
Ans. (a) Overfitting
5. Answer any 5 out of the given 6 questions (1 × 5 = 5 marks)
(i) Sonia developed a system that can understand and respond to spoken commands in various
languages. The system can translate and respond appropriately to user queries. Identify the
domain of AI in the given scenario.

(c) Data Science

S
(a) Natural Language Processing (b) Computer Vision
(d) Robotics
ER
Ans. (a) Natural Language Processing
(ii) Which evaluation metric calculates the proportion of true positive predictions out of all actual
positive instances?
(a) Accuracy (b) Precision
H

(c) Recall (d) F1 score

Ans. (c) Recall
(iii) ________________ describes the ability of a camera or image sensor to capture fine details and
OT

reproduce them accurately.

(a) Aperture (b) Exposure
(c) Sharpness (d) ISO
Ans. (c) Sharpness
BR

(iv) What does semantics of a language refer to?

(a) The meaning of words and sentences (b) The grammatical structure of sentences
(c) The pronunciation of words (d) The spelling of words
Ans. (a) The meaning of words and sentences
(v) Which technique is used to convert a collection of text into a matrix of token counts, helping
L

to analyze the frequency of words?

(a) Named Entity Recognition (b) Tokenisation
YA

(c) Bag of Words (d) Text Classification

Ans. (c) Bag of Words
(vi) In which scenario is minimizing false positives crucial due to high costs associated with
incorrect predictions?
GO

(a) Traffic congestion monitoring (b) Fraud detection in financial transactions

(c) Inventory management (d) Sentiment analysis in social media
Ans. (b) Fraud detection in financial transactions

Artificial Intelligence Class-10 (Subject Code-417) 5

 AN
SECTION B: OBJECTIVE TYPE QUESTIONS
Answer any 3 out of the given 5 questions on Employability Skills (2 × 3 = 6 marks)
Answer each question in 20 – 30 words.

H
6. Describe two best practices for effective communication.
Ans. Effective communication in a professional setting involves using active listening to fully understand

AS
the speaker's message and employing clear, concise language to avoid misunderstandings. Active
listening ensures that all parties are on the same page, while clarity in language helps in conveying
messages accurately, reducing the likelihood of confusion and improving overall productivity.
7. How does regular self-assessment contribute to effective self-management and personal

AK
development?
Ans. Regular self-assessment fosters self-awareness and accountability, helping individuals identify
strengths and areas for improvement. This continuous evaluation enables better goal setting,
decision-making, and personal growth.

PR
8. What is a “phishing attack” in the context of computer security.
Ans. A phishing attack involves deceptive emails or messages that trick users into revealing sensitive
information, like passwords or financial details, by pretending to be from a trustworthy source.
9. How do government grants and subsidies support entrepreneurs in starting and expanding
their businesses?
S
Ans. Government grants and subsidies provide financial assistance to entrepreneurs, reducing startup
costs and enabling business expansion. This support helps them overcome financial barriers and
ER
invest in growth and innovation.
10. How does excessive use of plastic contribute to environmental issues?
Ans. Excessive plastic use results in widespread pollution, as plastics do not decompose easily. They
accumulate in landfills, oceans, and waterways, harming wildlife, disrupting marine ecosystems, and
H

contributing to long-term environmental contamination.

Answer any 4 out of the given 6 questions in 20 – 30 words each (2 × 4 = 8 marks)
OT

11. What is the purpose of apps requesting access to camera and microphone?
Ans. Apps request access to the camera and microphone to enable features such as taking photos
or videos, recording audio, and participating in video calls. This access supports interactive
functionalities and content creation within the app.
BR

12. If Prisha and Disha aim to group data without predefined labels into different categories,
which learning model is most suitable, and what is its primary function?
Ans. They should use a clustering model, such as K-means or Gaussian Mixture Models. These models
perform unsupervised learning by analyzing feature patterns and grouping similar data points into
clusters, making them ideal for organizing unlabeled data.
L

13. What are two critical aspects Ajay must consider while accessing data from various sources
to maintain data integrity and usability?
YA

Ans. Ajay should focus on data consistency to ensure that the data across sources aligns correctly.
Additionally, he should check access permissions and rights to ensure he has the proper authorization
to use the data without legal or ethical issues.
14. Explain the term 'resolution' in the context of digital images with an example.
GO

Ans. Resolution refers to the clarity and detail of an image, defined by the number of pixels. For example,
a 1920 × 1080 resolution means the image has 1920 pixels horizontally and 1080 pixels vertically,
resulting in a clearer picture compared to a 640 × 480 resolution.

6 Model Test Paper 3 (Solved)

 AN
15. Identify any two stop words from the following sentence that should not be removed and
explain why they are essential: “Join us for an exciting event on June 5th at our new venue.”
Ans. The stop words "us" and "at" should not be removed. "Us" is crucial for indicating the recipient of

H
the action, while "at" provides location context. Removing these words would alter the sentence's
meaning and coherence.

AS
16. Draw the confusion matrix of the following data:
• The number of true positives = 100
• The number of true negatives = 47
• The number of false positives = 62

AK
• The number of false negatives = 290
Ans. The confusion matrix is as follows:
Predicted Positive Predicted Negative
Actual Positive 100 290

PR
Actual Negative 62 47
Answer any 3 out of the given 5 questions in 50 – 80 words each (4 × 3 = 12 marks)
17. Define artificial intelligence (AI) and describe with an example how it can be used to make
machines smarter.

S
Ans. Artificial Intelligence (AI) is the science of designing and developing systems that can mimic human
thinking and behavior to solve problems or perform tasks. AI enables machines to learn from
ER
experience and make decisions based on data. For example, in a self-driving car, AI algorithms
process data from sensors and cameras to understand the vehicle's surroundings, make decisions
about navigation, and react to obstacles. Over time, the system learns from each driving experience
to improve its safety and efficiency, demonstrating how machines can become more intelligent
H

through AI.
18. Rajan is learning about AI project Cycle. Explain the following to him:
• How can he apply the 4W Problem Canvas?
OT

• How can he apply Problem Statement Template to effectively scope his AI project?
Ans. Rajan can use the 4W Problem Canvas to capture a comprehensive view of the problem by addressing
the Who, What, Where, and Why aspects. This helps in understanding the context and significance
of the problem. Subsequently, the Problem Statement Template allows him to articulate the problem
BR

clearly by describing it, identifying existing challenges, and outlining the desired outcome. By
integrating insights from both tools, Rajan can ensure that his AI project is well-scoped, targeted,
and aligned with the needs of stakeholders, leading to more effective and relevant solutions.
19. Identify the type of learning in the following scenarios:
Scenario 1: A data analyst provides an AI model with a large amount of customer transaction
L

data without labels. The model then identifies patterns and groups customers based on their
purchasing behavior.
YA

Scenario 2: A model is trained using a dataset of labeled emails as either "spam" or "not
spam." After training, the model can classify new incoming emails based on what it learned
from the labeled data.
Ans. Scenario 1: This is an example of Unsupervised Learning, where the model identifies patterns or
GO

clusters in unlabeled data without being explicitly told what the labels should be.
Scenario 2: This is an example of Supervised Learning, where the model learns from labeled data
and applies that knowledge to classify new, unseen data.

Artificial Intelligence Class-10 (Subject Code-417) 7

 AN
20. Why is it challenging for computers to interact in human languages, and what are some
potential consequences of these challenges for applications like virtual assistants or
chatbots?

H
Ans. Computers face significant challenges when interacting in human languages due to the inherent
complexity and variability of language. Human languages are rich with nuances, context-dependent

AS
meanings, and cultural subtleties that are difficult for machines to grasp. For applications like virtual
assistants or chatbots, these challenges can lead to issues such as misinterpretation of user queries,
irrelevant or incorrect responses, and an inability to handle diverse linguistic inputs effectively. As
a result, users might experience frustration or dissatisfaction with these applications. To overcome

AK
these challenges, ongoing improvements in NLP technologies and continuous training on diverse
datasets are necessary to enhance the accuracy and effectiveness of language-based interactions in
AI systems.
21. Suppose that the model has to distinguish between smartphones and landline phones.

PR
Assume the smartphones to be the positive cases and landline phones the negative cases. Of
the 24 photos of smartphones, it correctly identifies 21 of them and 3 of them as landline
phones. Of the 18 photos of landline phones, it correctly identifies 15 of them and 3 of them
as smartphones.
Calculate the accuracy, precision and recall for the AI model.

S
Ans. (a) Accuracy = Number of correct predictions/Number of predictions made
= (21 + 15)/(24 + 18) = 36/42 = 0.86
ER
(b) Precision = True Positives/Total number of positives
= True Positive/(True positives + False positives) = 21/21 + 3 = 0.875
(c) Recall = True Positives/(True positives + False negatives)
H

= 21/(21 + 3) [Since it calls 3 of smartphone photos as landline, wrongly]

= 0.875
OT
BR
L
YA
GO

8 Model Test Paper 3 (Solved)