ARTIFICIAL INTELLIGENCE (SUBJECT CODE - 417)

Class X (2023-24)
Sample Paper -2

Max. Time: 2 Hours Max. Marks: 50

General Instructions:

1. Please read the instructions carefully.

2. This Question Paper consists of 21 questions in two sections: Section A & Section B.
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.
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.
7. SECTION B- SUBJECTIVE TYPE QUESTIONS (26 MARKS):
ps
i. This section has 16 questions.
ii. A candidate has to do 10 questions.
iii. Do as per the instructions given.
iv. Marks allotted are mentioned against each question/part.
Ki

© Kips Learning Pvt. Ltd 2024

 SECTION A: OBJECTIVE TYPE QUESTIONS
Q. 1 Answer any 5 out of the given 6 questions (1 x 5 = 5 marks)
i. Which of the following is not an element to keep a person motivated? 1
a. Commitment to personal or organisational goal
b. Initiative or readiness to act on opportunities
c. Being aware of your personality
d. Being socially active
ii. Which one of the following is a correct statement? 1
a. All files are of the same type.
b. Only files of the same type can be stored in a folder.
c. Files can be of different types.
d. All the deleted files go to the Recycle bin.
iii. Assertion (A): Entrepreneurs play a crucial role in economic development. 1
Reason (R): Entrepreneurs focus primarily on personal gain and often hinder overall
economic growth.
a. Both Assertion and Reason are true, and Reason is the correct explanation of
Assertion.
b. Both Assertion and Reason are true, but Reason is not the correct explanation of
Assertion.

iv.
ps
c. Assertion is true, but Reason is false.
d. Assertion is false, but Reason is true.
What is the main objective of Sustainable Development Goal 17?
a. Eliminate poverty
b. Ensure gender equality
c. Promote peace, justice, and strong institutions
d. Strengthen the global partnership for sustainable development
1

v. Suresh, a team member, is proficient in Hindi but faces challenges in expressing complex 1
technical details due to difficulties with English. What type of communication barrier
does Suresh face?
Ki
a. Linguistic barrier
b. Physical barrier
c. Semantic barrier
d. Psychological barrier
vi. Recently, Smita lost her part-time job, and she is struggling to cover her living expenses. 1
What type of stress is she likely experiencing?
a. Academic stress
b. Financial stress
c. Social stress
d. Time-related stress

Q. 2 Answer any 4 out of the given 6 questions on Employability Skills (1 x 4 = 4 marks)

i. Assertion (A): A person might be good at drawing but bad at math. 1
Reason (R): Drawing skills and math skills are learned differently and use different parts
of the brain.
a. Both Assertion and Reason are true, and Reason is the correct explanation of
Assertion.

© Kips Learning Pvt. Ltd 2024

 b. Both Assertion and Reason are true, but Reason is not the correct explanation of
Assertion.
c. Assertion is true, but Reason is false.
d. Assertion is false, but Reason is true.
ii. A healthcare provider implemented an AI system to prioritise patient treatment 1
schedules. Over time, the system began to prioritise patients with higher incomes for
earlier appointments, irrespective of medical urgency, causing a debate about equitable
access to healthcare.
What is the central issue in this scenario?
a. Cloud Computing
b. System Scalability
c. Data Mining
d. AI Ethics
iii. Statement 1: Deep learning algorithms can improve their accuracy by learning from their 1
mistakes when additional layers are added to the network.
Statement 2: Conventional algorithms are always capable of handling high-dimensional
data without any increase in computational complexity.
a. Both Statement1 and Statement2 are correct
b. Both Statement1 and Statement2 are incorrect
c. Statement1 is correct but Statement2 is incorrect
ps
d. Statement2 is correct but Statement1 is incorrect
iv. Given a company that wants to build a machine learning system with scalable performance 1
and has access to a moderate budget for compute capacity, which approach should the
company consider based on the provided graph?
Ki

a. Rely solely on traditional AI/ML models since they work effectively on CPUs and offer
decent performance.
b. Start with a medium neural network and invest in some accelerators for improved
compute capacity.
c. Choose a small neural network and only invest in CPUs to manage costs.
d. Directly adopt a large neural network and heavily invest in accelerators for maximal
algorithmic quality.
v. ___________ is a type of artificial intelligence that is programmed to mimic human 1
conversation through text or spoken words, often used in customer service.
a. Speech Recognition
b. Chatbot

© Kips Learning Pvt. Ltd 2024

 c. Recommendation System
d. Facial Recognition
vi. In a series of 8 blog posts discussing artificial intelligence, the phrase 'neural pruning' is 1
mentioned just three times and solely in two of the posts. How would you categorize the
phrase 'neural pruning' in terms of its frequency of appearance?
a. Stop word
b. Rare word
c. Common word
d. Irrelevant word

Q. 3 Answer any 5 out of the given 6 questions (1 x 5 = 5 marks)

i. Consider the following examples: 1
i. A thermostat adjusting the temperature based on the room's occupancy.
ii. An app that translates spoken language in real-time.
iii. A watch that measures steps and calories burned.
iv. A social media platform that filters out spam messages automatically.
Choose the options that are not AI:
a. i and ii
b. i and iii
ps
c. ii and iv
d. iii and iv
ii. During data acquisition, feeding previous data is called _____________. 1
a. Training data
b. Testing data
c. Evaluating data
d. Predicting data
iii. Which of the following does NOT typically involve the use of data science? 1
Ki
a. Optimizing delivery routes for logistics companies
b. Generating virtual assistants to handle customer service
c. Creating digital animations for movies
d. Diagnosing diseases from medical imaging
iv. _________ is a method used to detect the edges within an image, which helps in 1
understanding the shape and features of different objects in the image.
a. Edge detection
b. Colour filtering
c. Pattern recognition
d. Feature extraction
v. Determine the category of the following Chatbot: 1
This chatbot can conduct a conversation in natural language and learn from each
interaction, adjusting its responses to improve accuracy over time. It uses sophisticated
algorithms to recognize speech or text inputs and can often handle a wide range of
topics without predefined scripts.
vi. In a scenario where a diagnostic tool must minimize the rate of false negatives while 1
ensuring that the positive detections are truly relevant, which pair of evaluation metrics
becomes particularly important?
a. Recall and Specificity

© Kips Learning Pvt. Ltd 2024

 b. Accuracy and Precision
c. Recall and Precision
d. Specificity and Accuracy

Q. 4 Answer any 5 out of the given 6 questions (1 x 5 = 5 marks)

i. Vandana was in a discussion about modern technology and remembered reading about a 1
type of network that simulates the human brain to process data. She is having a hard
time recalling the specific term for this kind of network. Can you provide her with the
right term?
ii. Statement 1: In deep learning, a convolutional neural network is specifically designed to 1
process pixel data.
Statement 2: In supervised learning, ‘labeled data’ means that each sample in the
dataset is paired with a noise signal to enhance model robustness.
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
ps
d. Statement 2 is correct but Statement 1 is incorrect
iii. Aarav is working on a neural network that classifies emails as spam or not spam by 1
analysing the text within the emails. His network has been trained using a substantial
amount of labelled data and has shown impressive precision and accuracy rates. Which
of the following statements is not true?
a. The labelled dataset containing examples of spam and not spam emails is considered
the Training Data.
b. Precision and accuracy rates are performance metrics that are calculated using the
Validation Data.
Ki
c. After training, the neural network's performance is assessed on a separate set of
emails known as Test Data.
d. In an ideal scenario, half of the available data should be used for training, and the
other half for validation and testing.
iv. Which deep learning architecture is commonly used for image classification tasks? 1
a. Recurrent Neural Network (RNN)
b. Convolutional Neural Network (CNN)
c. Generative Adversarial Network (GAN)
d. Long Short-Term Memory (LSTM)
v. Which NLP method is often utilized to retrieve structured information from unstructured 1
text data, such as extracting company names from news articles?
a. Text Summarisation
b. Sentiment Analysis
c. Part-of-Speech Tagging
d. Named Entity Recognition
vi. Aman built a language translation model that could flawlessly translate sentences from 1
English to French using a dataset of classic literature texts. He carefully ensured that the
translations were accurate for each text in his training set. Excited by the model's
performance on these texts, he presented it to his peers. But then, he found that the
model struggled to translate modern slang and casual phrases used in today’s

© Kips Learning Pvt. Ltd 2024

 conversations, which it did not learn from his classic literature training data. Name the
concept.

Q. 5 Answer any 5 out of the given 6 questions (1 x 5 = 5 marks)

i. Sara, a software developer, has created an application that transcribes spoken words into 1
written text in real-time during meetings. The application can recognise speech in various
accents and dialects and is designed to improve the accuracy of live captioning and note-
taking. She has incorporated a vast library of spoken language samples to enhance the
system's learning. Which AI domain does Sara's application fall under?
a. Robotics
b. Natural Language Processing
c. Machine Learning
d. None of these
ii. Which evaluation metric focuses on the proportion of actual positives that are correctly 1
identified?
iii. ______ is the term used to describe the sharpness or clarity of an image, often determined 1

iv.
c. Clarity
ps
by the number of pixels in a specific area.
a. Definition
b. Resolution

d. Pixilation
Which of the following best describes the role of syntax in a language? 1
a. Deciphering the tone of a statement
b. Understanding the cultural context of a word
c. Dictating the proper arrangement of words in a sentence
Ki
d. Interpreting the emotions behind a sentence
v. Which text processing technique involves breaking a text into individual words or terms, 1
often to be used as input for other methods like frequency counts or text modelling?
a. Lemmatization
b. Stopword Removal
c. Tokenization
d. Sentence segmentation
vi. Which one of the following scenarios results in a high false positive cost? 1
a. Metal detector at an airport
b. Predicting rain for agricultural crops
c. Earthquake warning system
d. A movie recommendation system

SECTION B: SUBJECTIVE TYPE QUESTIONS

Answer any 3 out of the given 5 questions on Employability Skills (2 x 3 = 6 marks)
Answer each question in 20 – 30 words.
Q. 6 What is the importance of feedback? 2
Q. 7 Write any two stress management techniques. 2

© Kips Learning Pvt. Ltd 2024

 Q. 8 Anand wants to make sure his computer is safe from viruses. What are two key 2
strategies he can implement to ensure effective protection?
Q. 9 What is the relationship between an entrepreneur and society? 2
Q. 10 List any two key concepts of sustainable development. 2
Answer any 4 out of the given 6 questions in 20 – 30 words each (2 x 4 = 8 marks)

Q. 11 In our daily lives, we often engage with social media. On these platforms, our feeds are 2
usually populated with posts and ads that seem tailored to our interests. How do social
media platforms curate such personalized content?
Q. 12 Priya and Navin want to design a system that can categorize news articles into different 2
topics without being provided with any predefined categories. Which type of learning
model should they implement and why?
Q. 13 What are two key factors that need to be kept in mind while collecting data from 2
different data sources?
Q. 14 Differentiate between the grayscale and RGB images. 2
Q. 15 Identify any two stop words which should not be removed from the given sentence and 2
why?
‘Even if the weather looks bad, we can still go out for dinner, unless you’d prefer to stay
in.’ ps
Q. 16 Construct the confusion matrix based on the given information: 2
• the number of true positive = 125
• the number of true negative = 80
• the number of false positive = 45
• the number of false negative = 150
Answer any 3 out of the given 5 questions in 50– 80 words each (4 x 3 = 12 marks)

Q. 17 During a family dinner, your aunt asks you about your work in artificial intelligence, 4
specifically in neural networks. She is intrigued but doesn't quite understand how they
Ki
function. Explain to her what a neural network is and give a relatable example of how
neural networks can be used to solve complex problems.
Q. 18 Maya is tasked with overseeing the development of a new machine learning model for her 4
company. Clarify the following for her:
a. Training Dataset
b. Algorithm Selection Criteria

© Kips Learning Pvt. Ltd 2024

Q. 19 If a machine learning algorithm were to categorize songs into different genres after being 4
trained on Dataset X, would it be capable of distinguishing between Rock and Jazz when
exposed to new songs? What kind of learning is illustrated by Dataset X, and how does it
differ from the learning process that Dataset Y would require?

Figure A (Dataset X) Figure B (Dataset Y)

Q. 20 Autonomous vehicles rely on precise algorithms and sensor data to navigate safely. They
ps 4
do not possess human-like judgement or ethical reasoning. Do you believe that self-
driving cars might encounter obstacles when faced with moral dilemmas on the road, such
as those proposed in the trolley problem? Discuss your perspective.
Q. 21 A diagnostic tool was used to determine whether patients were at risk for a certain heart 4
condition:
Reality
Confusion Matrix At Risk Not at Risk
At Risk 95 30
Ki
Prediction Not at Risk 20 55

i. Compute the total number of errors made by the diagnostic tool.

ii. Determine precision, recall, and the F1 Score for the ‘At Risk’ classification.

© Kips Learning Pvt. Ltd 2024