Class X - final -Question Paper

Artificial Intelligence
Note:

All questions are compulsory

Section-A

Q1 [4x1=4]
i. What does the term "sustainable development" means?

A. Economic development at all costs

B. Development that does not harm the environment
C. Using more natural resources
D. Technological progress without limits

ii. how do entrepreneurs help grow the area and society they live in?

A. By keeping all profits to themselves

B. By avoiding the market system
C. By creating opportunities for buying and selling, which helps everyone earn money
D. By reducing the number of sellers in the market

iii. One advantage of visual communication is that it stays ______ across

different places, making it easy to understand globally.

Iv. To protect data from being compromised by malicious software, one

effective preventive measure is installing ______ that can detect,
quarantine, and eliminate threats before they cause harm.

Q2 [5x1=5]

I. Which of the following best describes Deep Learning (DL)?

A. A method where programmers write specific rules for every task

B. A process where machines manually collect data from users
C. A form of AI where machines train themselves using large amounts of data to create their
own algorithms
D. A basic software tool used only for storing information
ii. What is the purpose of a testing data set in machine learning?

A. To train the model with new information

B. To improve the speed of the model
C. To evaluate how accurately the model performs on unseen data
D. To store data permanently

iii. What is the main purpose of frameworks in problem-solving?

A. To confuse people with complex steps

B. To provide a step-by-step, organized, and structured approach for solving problems
C. To replace the need for communication between team members
D. To solve problems randomly without any plan

Iv. In the Learning Based approach to AI, the model is trained on data and
can ______ itself to handle changes or exceptions in new data.

V. The ethical framework that focuses on the character and intentions of

individuals, emphasizing honesty, compassion, and integrity, is called
______.

Q3 [5x1=5]

I. How do search engines like Google deliver the best results for our
queries in a fraction of a second?

A. By using data science algorithms to process vast amounts of data quickly

B. By manually reviewing each search query before providing results
C. By limiting the number of search results to improve speed
D. By relying solely on user feedback for search rankings

ii. What is a key feature of No-Code AI tools?

A. They require advanced coding skills

B. They allow users to drag and drop models without coding
C. They only work with simple algorithms
D. They need manual programming for every step

iii. The ______ dataset is used to train the model, while the ______ dataset is
used to test the model’s predictions by comparing them to expected
values.
Iv. Accuracy is an evaluation metric that measures the total number of
correct ______ made by a model, indicating its performance.

V. The process of sorting items into different groups, such as placing fruits
and vegetables in one trolley and groceries in another, is called ______.

Q4 [5x1=5]

I. What is the main purpose of computer vision?

A. To create video games

B. To extract and understand information from images or videos
C. To write computer programs
D. To store large amounts of data

ii. How do face filters in apps like Instagram and Snapchat work using
computer vision?

A. By manually selecting and applying filters to photos after they are taken
B. By detecting and analyzing facial dynamics in real-time to apply selected filters
C. By storing pre-edited images and swapping faces automatically
D. By using audio signals to trigger filter effects

iii. Which of the following is NOT a way Computer Vision is used in retail
according to the passage?

A. Analyzing customer navigation routes in the store

B. Detecting walking patterns of customers
C. Tracking customers’ movements through stores
D. Automatically restocking shelves without human input

Iv. Computer-supported medical imaging helps doctors by converting 2D

scan images into interactive 3D ______, providing a detailed understanding
of a patient’s health.

V. The Google Translate app uses ______ character recognition to read text
from images and augmented reality to display translations instantly.

Q5 [5x1=5]
i. Why is Natural Language Processing (NLP) important in computer
communication?

A. It helps computers understand programming languages like Python and Java

B. It converts human natural language into a form that computers can understand
C. It allows computers to generate images from text descriptions
D. It stores data for faster retrieval

ii. How do voice assistants use Natural Language Processing (NLP)?

A. By converting text messages into speech

B. By understanding natural speech to execute tasks efficiently
C. By translating foreign languages into English
D. By generating images based on voice commands

iii. How can keyword extraction tools enhance customer service in

businesses?

A. By generating automated responses without analyzing customer feedback

B. By identifying frequently mentioned words and sentiments in social media, enabling
businesses to address customer concerns proactively
C. By summarizing legal documents for internal review
D. By translating customer complaints into different languages for global outreach

Iv. ______ is the process in NLP that involves dividing a large chunk of text
into paragraphs, sentences, and words to identify the structure of input
words.

V. When working with text from multiple documents altogether, the entire
collection of textual data is called a ______.

Section-B [7x2=14]

Q6 Who are online predators, and what are two common tactics they use to
exploit others?

Q7 Read the case below and answer the questions that follow:

Riya wants to start a small food stall in her neighborhood. She notices there are already many
food stalls, but none of them offer healthy meal options for office-goers. She plans to serve
affordable, nutritious meals using organic ingredients.
What common misconception about business ideas is challenged in this case? How is Riya
applying a new idea to an existing business model?

Q8 In a school group project, some students only want to work with their close friends and
ignore a few classmates who come from different backgrounds. However, Ayaan suggests that
everyone should be included, regardless of their background, and ensures that each member
gets a chance to contribute.

Which principle of reducing inequalities is Ayaan following?Mention one way students can
promote equality during group activities.

Q9 Riya is working on analyzing customer reviews for her company. While processing the text
data, she notices many common words like "is," "the," and "and" appearing frequently. She
decides to remove these words to focus on more meaningful content.

What are these commonly removed words called in text processing? Why is it important to
remove these words during text analysis?

Q10 Read the passage below and answer the following question:

The word "pixel" refers to a picture element, which is the smallest unit of a digital image. Pixels
are arranged in a 2D grid and together form the complete image. When magnified, individual
pixels become visible, and the level of detail in an image depends on the number of pixels used.
Higher pixel count results in better image clarity and resemblance to the original.

Why does increasing the number of pixels in a digital image improve its quality? Explain using
your understanding of pixel arrangement and image approximation.

Q11 Problem Context: Many college students struggle to stay focused during online classes.

Prepare 4W Problem Canvas for the Problem Context given above.

Q12 Observe the picture given below carefully and answer the following question

What activity is taking place in the picture? Which concept of AI is being represented
here?
Section- C [3x4=12]

Q13 The selection committee has a record of 15 students for a tournament. The committee
recommended some students, and later the final selection result (whether the student was
actually selected or not) was recorded as follows:
Student Recommended by Actually
Committee Selected

A Yes No

B Yes Yes

C No Yes

D Yes Yes

E Yes No

F No No

G Yes Yes

H No No

I Yes Yes
J No No

K No Yes

L Yes No

M No No

N Yes Yes

O No No

Tasks:

1. Fill in the confusion matrix for the recommendation vs actual selection.

2. Calculate the Accuracy of the committee's recommendation.

3. Calculate the Precision of the recommendation.

4. Calculate the Recall (Sensitivity) of the recommendation.

Q14 Consider the following three documents:

Document ID Text

Doc 1 "Artificial intelligence is the future"

Doc 2 "The future of AI is bright"

Doc 3 "Many companies invest in artificial intelligence"

Calculate the Term Frequency (TF) and also calculate the Inverse Document Frequency (IDF)

Q15

a) Read the information given below:

Image 1: Shows groups of different colored dots separated into clusters without predefined
labels.

Image 2: Shows dots labeled as different categories (e.g., apples vs oranges) based on known
classes.

Explain what clustering is and how it differs from classification based on these images.

b) Given the text:

“We are the students of SPSEC. SPSEC is a CBSE school.”

Perform the following tasks:

● Segment the text into sentences.

● Tokenize the sentences into words.

● Remove stop words from the tokens.

● Convert the tokens to lowercase.

● Provide the stemmed and lemmatized forms of the remaining tokens.

=========================
 Class X - final - answer key
Artificial Intelligence
Syllabus:

Employability Skills
Unit 1: Communication Skills-II
Unit 2: Self-Management Skills-II
Unit 3: ICT Skills-II
Unit 4: Entrepreneurial Skills-II
Unit 5: Green Skills-II

Subject Specific Skills Theory

Unit 1: Revisiting AI Project Cycle & Ethical Frameworks for AI
Unit 2: Advanced Concepts of Modeling in AI
Unit 3: Evaluating Models
Unit 4: Statistical Data
Unit 5: Computer Vision
Unit 6: Natural Language Processing

All questions are compulsory

Section-A DONE

Q1 empl [4x1=4]
I. mcq

What does the term "sustainable development" means?

A. Economic development at all costs

B. Development that does not harm the environment
C. Using more natural resources
D. Technological progress without limits

Correct Answer: B. Development that does not harm the environment

ii. mcq

how do entrepreneurs help grow the area and society they live in?
A. By keeping all profits to themselves
B. By avoiding the market system
C. By creating opportunities for buying and selling, which helps everyone earn money
D. By reducing the number of sellers in the market

Correct Answer: C. By creating opportunities for buying and selling, which helps everyone earn
money

iii. fib

One advantage of visual communication is that it stays ______ across

different places, making it easy to understand globally.

Answer: consistent

Iv. fib

To protect data from being compromised by malicious software, one

effective preventive measure is installing ______ that can detect,
quarantine, and eliminate threats before they cause harm.

Answer: anti-virus software

Q2 proj cycle and adv concpt [5x1=5]

I. mcq

Which of the following best describes Deep Learning (DL)?

A. A method where programmers write specific rules for every task

B. A process where machines manually collect data from users
C. A form of AI where machines train themselves using large amounts of data to create their
own algorithms
D. A basic software tool used only for storing information

Correct Answer: C. A form of AI where machines train themselves using large

amounts of data to create their own algorithms

ii. Mcq
What is the purpose of a testing data set in machine learning?

A. To train the model with new information

B. To improve the speed of the model
C. To evaluate how accurately the model performs on unseen data
D. To store data permanently

Correct Answer:
C. To evaluate how accurately the model performs on unseen data

iii. mcq

What is the main purpose of frameworks in problem-solving?

A. To confuse people with complex steps

B. To provide a step-by-step, organized, and structured approach for solving problems
C. To replace the need for communication between team members
D. To solve problems randomly without any plan

Correct Answer:
B. To provide a step-by-step, organized, and structured approach for solving problems

Iv. fib

In the Learning Based approach to AI, the model is trained on data and can
______ itself to handle changes or exceptions in new data.

Answer: modify

V. fib

The ethical framework that focuses on the character and intentions of

individuals, emphasizing honesty, compassion, and integrity, is called
______.

Answer: virtue ethics

Q3 eval FIB and stats data MCQ [5x1=5]

I. mcq

How do search engines like Google deliver the best results for our queries
in a fraction of a second?

A. By using data science algorithms to process vast amounts of data quickly

B. By manually reviewing each search query before providing results
C. By limiting the number of search results to improve speed
D. By relying solely on user feedback for search rankings

Correct Answer:
A. By using data science algorithms to process vast amounts of data quickly

ii. mcq

What is a key feature of No-Code AI tools?

A. They require advanced coding skills

B. They allow users to drag and drop models without coding
C. They only work with simple algorithms
D. They need manual programming for every step

Correct Answer:
B. They allow users to drag and drop models without coding

iii. fib

The ______ dataset is used to train the model, while the ______ dataset is
used to test the model’s predictions by comparing them to expected
values.

Answer: train, test

Iv. fib

Accuracy is an evaluation metric that measures the total number of correct

______ made by a model, indicating its performance.

Answer: predictions
V. fib

The process of sorting items into different groups, such as placing fruits
and vegetables in one trolley and groceries in another, is called ______.

Answer: classification

Q4 CV [5x1=5]

I. mcq

What is the main purpose of computer vision?

A. To create video games

B. To extract and understand information from images or videos
C. To write computer programs
D. To store large amounts of data

Correct Answer:
B. To extract and understand information from images or videos

ii. mcq

How do face filters in apps like Instagram and Snapchat work using
computer vision?

A. By manually selecting and applying filters to photos after they are taken
B. By detecting and analyzing facial dynamics in real-time to apply selected filters
C. By storing pre-edited images and swapping faces automatically
D. By using audio signals to trigger filter effects

Correct Answer:
B. By detecting and analyzing facial dynamics in real-time to apply selected filters

iii. mcq

Which of the following is NOT a way Computer Vision is used in retail

according to the passage?
A. Analyzing customer navigation routes in the store
B. Detecting walking patterns of customers
C. Tracking customers’ movements through stores
D. Automatically restocking shelves without human input

Correct Answer:
D. Automatically restocking shelves without human input

Iv. fib

Computer-supported medical imaging helps doctors by converting 2D scan

images into interactive 3D ______, providing a detailed understanding of a
patient’s health.

Answer: models

V. fib

The Google Translate app uses ______ character recognition to read text
from images and augmented reality to display translations instantly.

Answer: optical

Q5 NLP [5x1=5]

I. mcq

Why is Natural Language Processing (NLP) important in computer

communication?

A. It helps computers understand programming languages like Python and Java

B. It converts human natural language into a form that computers can understand
C. It allows computers to generate images from text descriptions
D. It stores data for faster retrieval

Correct Answer:
B. It converts human natural language into a form that computers can understand

ii. mcq
How do voice assistants use Natural Language Processing (NLP)?

A. By converting text messages into speech

B. By understanding natural speech to execute tasks efficiently
C. By translating foreign languages into English
D. By generating images based on voice commands

Correct Answer:
B. By understanding natural speech to execute tasks efficiently

iii. mcq

How can keyword extraction tools enhance customer service in

businesses?

A. By generating automated responses without analyzing customer feedback

B. By identifying frequently mentioned words and sentiments in social media, enabling
businesses to address customer concerns proactively
C. By summarizing legal documents for internal review
D. By translating customer complaints into different languages for global outreach

Correct Answer:
B. By identifying frequently mentioned words and sentiments in social media, enabling
businesses to address customer concerns proactively

Iv. fib

______ is the process in NLP that involves dividing a large chunk of text
into paragraphs, sentences, and words to identify the structure of input
words.

Answer: Lexical Analysis

V. fib

When working with text from multiple documents altogether, the entire
collection of textual data is called a ______.

Answer: corpus
Section-B [7x2=14]

Q6 ICT

Who are online predators, and what are two common tactics they use to
exploit others?

Answer :
Online predators are individuals who manipulate others into inappropriate relationships. Two
common tactics they use include pretending to be someone their victim’s age and pressuring
them into illegal activities, either online or in person.

Q7 ENTER CASE BASED

Read the case below and answer the questions that follow:

Case:
Riya wants to start a small food stall in her neighborhood. She notices there are already many
food stalls, but none of them offer healthy meal options for office-goers. She plans to serve
affordable, nutritious meals using organic ingredients.

What common misconception about business ideas is challenged in this case? How is Riya
applying a new idea to an existing business model?

Answers:

The misconception that every business idea must be unique or special. Riya is offering a
healthy and affordable food option, which is different from what existing stalls provide, thereby
meeting a specific customer demand.

Q8 GREEN SK

Case:
In a school group project, some students only want to work with their close friends and ignore a
few classmates who come from different backgrounds. However, Ayaan suggests that everyone
should be included, regardless of their background, and ensures that each member gets a
chance to contribute.

Which principle of reducing inequalities is Ayaan following?Mention one way students can
promote equality during group activities.

Answers:
Ayaan is following the principle of including everyone while working or playing, regardless of
their background. By being friendly and helpful to all classmates and giving equal
opportunities to participate, students can promote equality.

Q9 NLP - case based

Case:

Riya is working on analyzing customer reviews for her company. While processing the text data,
she notices many common words like "is," "the," and "and" appearing frequently. She decides to
remove these words to focus on more meaningful content.

What are these commonly removed words called in text processing? Why is it important to
remove these words during text analysis?

Answers:
These words are called stop words. They are removed because they occur very frequently but
do not add valuable meaning to the analysis.

Q10 CV - case based

Read the passage below and answer the following question:

The word "pixel" refers to a picture element, which is the smallest unit of a digital image. Pixels
are arranged in a 2D grid and together form the complete image. When magnified, individual
pixels become visible, and the level of detail in an image depends on the number of pixels used.
Higher pixel count results in better image clarity and resemblance to the original.

Why does increasing the number of pixels in a digital image improve its quality? Explain
using your understanding of pixel arrangement and image approximation.

✅ Answer
Increasing the number of pixels improves image quality because more pixels provide finer
details and smoother transitions between colors and shapes. Since pixels approximate the
original image, having more of them allows the image to more closely resemble the real object,
reducing blurriness and blockiness.

Q11 4w prob canvas LEFT

Problem Context: Many college students struggle to stay focused during online classes.

Prepare 4W Problem Canvas for the Problem Context given above.

Ans:

Using the 4W Problem Canvas, fill in the table:

W Answer (Example)

Who College students attending online classes

What Difficulty staying focused and engaged during lectures

Where During remote/online learning sessions

Why Leads to poor understanding, low performance, and

disengagement
Q12 adv concepts - picture based

Observe the picture given below carefully and answer the following questions:

What activity is taking place in the picture? Which concept of AI is being represented
here?

ANSWER:
Supervised learning is a type of machine learning where the model is trained on labeled data,
meaning the input comes with correct output. It learns to predict or classify based on this
training.
Unsupervised learning involves unlabeled data, where the model identifies patterns, clusters,
or groupings without predefined outputs.

Section- C DONE [3x4=12]

Q13 conf matrix

Question:

The selection committee has a record of 15 students for a tournament. The committee
recommended some students, and later the final selection result (whether the student was
actually selected or not) was recorded as follows:

Student Recommended by Actually

Committee Selected
A Yes No

B Yes Yes

C No Yes

D Yes Yes

E Yes No

F No No

G Yes Yes

H No No

I Yes Yes

J No No

K No Yes

L Yes No

M No No

N Yes Yes

O No No

Tasks:

5. Fill in the confusion matrix for the recommendation vs actual selection.

6. Calculate the Accuracy of the committee's recommendation.

 7. Calculate the Precision of the recommendation.

8. Calculate the Recall (Sensitivity) of the recommendation.

Answer:

1. Confusion Matrix:

Actually Selected Actually Not Selected

(Yes) (No)

Recommended (Yes) TP = 5 (B, D, G, I, N) FP = 4 (A, E, L)

Not Recommended (No) FN = 2 (C, K) TN = 4 (F, H, J, M, O)

Check counts:

● Recommended Yes & Selected Yes (TP): B, D, G, I, N → 5

● Recommended Yes & Selected No (FP): A, E, L → 3 (note: you wrote 4, but only 3 here)

● Recommended No & Selected Yes (FN): C, K → 2

● Recommended No & Selected No (TN): F, H, J, M, O → 5

(Counting carefully: Total students = 15)

2. Accuracy = ({TP + TN}{Total} = {5 + 5}{15} = {10}{15} = 0.6667 = 66.67%)

3. Precision = ({TP}{TP + FP} = {5}{5 + 3} = {5}{8} = 0.625 = 62.5%)

4. Recall = ({TP}{TP + FN} = {5}{5 + 2} = {5}{7} \approx 0.714 = 71.4%)

Q14 tf idf

Question:

Consider the following three documents:

Document ID Text

Doc 1 "Artificial intelligence is the future"

Doc 2 "The future of AI is bright"

Doc 3 "Many companies invest in artificial intelligence"

Calculate the Term Frequency (TF) and also calculate the Inverse Document Frequency (IDF)

Answer:

PROVIDE COMPLETE TABULAR FORMAT AS DISCUSSED IN THE CLASS

Certainly! Here’s a profound and clear tabular format for TF and TF-IDF calculations of the
word "future" in the given documents:

Docume Text Tota Frequency Term Inverse TF-

nt l of Frequ Document I
ID W "futur ency Frequenc D
o e" (TF) y (IDF) F
r (
d T
s F
×
I
D
F
)
Doc 1 Artificial 5 1 1/5 = 0.20 \multirow{3}{*}{l 0.20 ×
intelligenc 0
e is the og(3/2) ≈ .
future 0.176} 1
7
6
=
0
.
0
3
5
2

Doc 2 The future of AI 6 1 1/6 ≈ 0.167 0.167

is bright
×
0

.
1
7
6
≈
0
.

0
2

9
4

Doc 3 Many 6 0 0 0×
companie 0
s invest in .
artificial 1
intelligenc 7
e 6
=
0

Notes:
 ● TF (Term Frequency) is the frequency of the term in the document divided by total
words in that document.

● IDF (Inverse Document Frequency) is constant across documents for the term "future"
because it depends on the entire corpus.

● TF-IDF highlights the importance of the term "future" in each document.

Q15 CV

a. Cv - object detection and classification

Read the information given below:

Image 1: Shows groups of different colored dots separated into clusters without predefined
labels. Image 2: Shows dots labeled as different categories (e.g., apples vs oranges) based on
known classes.

Explain what clustering is and how it differs from classification based on these images.

Answer:

Clustering is grouping similar data points together without pre-labeled categories (unsupervised
learning). Classification assigns data points to predefined categories or labels based on training
data (supervised learning).

b. Nlp - normalization

Given the text:

“We are the students of SPSEC. SPSEC is a CBSE school.”

Perform the following tasks:

● Segment the text into sentences.

● Tokenize the sentences into words.

● Remove stop words from the tokens.

 ● Convert the tokens to lowercase.

● Provide the stemmed and lemmatized forms of the remaining tokens.

Here’s a brief 2-mark answer for the question:

Answer:

● Sentence Segmentation:

1. We are the students of SPSEC.

2. SPSEC is a CBSE school.

● Tokenization:

1. [We, are, the, students, of, SPSEC]

2. [SPSEC, is, a, CBSE, school]

● Remove Stop Words:

[students, SPSEC, SPSEC, CBSE, school]

● Convert to Lowercase:
[students, spsec, spsec, cbse, school]

● Stemming and Lemmatization:

Token Stemme Lemmatized

d

student student student

s

spsec spsec spsec

cbse cbse cbse

school school school