KENDRIYA VIDYALAYA SANGTHAN

LUCKNOW REGION
ARTIFICIAL INTELLIGENCE (SUBJECT CODE - 417)
MODEL ANSWERS SHEET CLASS X (SESSION 2024-2025)
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. There is no negative marking.
iii. Do as per the instructions given.
iv. Marks allotted are mentioned against each question/part.
7. SECTION B – SUBJECTIVE TYPE QUESTIONS (26 MARKS):
i. This section contains 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.

SECTION A: OBJECTIVE TYPE QUESTIONS

Q.
QUESTION Mark
No.
s
.
Q. 1 Answer any 4 out of the given 6 questions on Employability Skills (1 x 4 = 4 marks)
i. a) leadership and time management 1

ii. (c) Trojan hourse 1

iii. (a) Both A and R are true, and R is the correct explanation of A. 1

b) To adjust strategies and improve efficiency

iv. 1

v. (a) Language barrier 1

vi. (d) waste reduction 1

Q. 2 Answer any 5 out of the given 6 questions (1 x 5 = 5
marks)

417 - ARTIFICIAL INTELLIGENCE – X PAGE 1 OF 8

 i. (a) Both A and R are correct and R is the correct explanation of A 1

ii. b). Improving diagnostic accuracy 1

iii. (a) Both A and R are correct and R is the correct explanation of A 1

iv. c). Design and Development 1

v. d. All of the above 1

vi. (c) Stop word 1

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

i. a) Human intelligence can feel emotions; AI cannot. 1

ii. Training Data 1

iii. b). To prepare data for analysis by removing inaccuracies and inconsistencies 1

iv. a) Pixel 1

v. b). Tokenization 1

vi. b) The model incorrectly labels many non-spam emails as spam. 1

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

i. d) all of these 1

ii.
(d) Statement2 is correct but Statement1 is incorrect 1

 Statement1 is incorrect because, in AI, a high accuracy model is not always

preferred. The type of data and the context are critical. For instance, in cases with
class imbalance (like fraud detection), a high accuracy model might just be
capturing the majority class, missing critical insights.
 Statement2 is correct, as accuracy alone can indeed be misleading. For example,
in an imbalanced dataset where one class dominates, a model that predicts only
the majority class will have high accuracy but poor real-world utility.

The correct answer is:

iii. 1
(a) Testing data is used to train the model to learn the relationships between
weather, soil, and crop health.

 This statement is incorrect because testing data is not used for training the model.
Instead, training data is used to help the model learn relationships and patterns,
while testing data is set aside to evaluate the model’s performance after training.

417 - ARTIFICIAL INTELLIGENCE – X PAGE 2 OF 8

 iv. Answer: RGB 1

v.
(b) Speech Recognition 1

 Speech Recognition is the NLP feature that allows virtual assistants like Siri and
Alexa to understand and process spoken commands by converting spoken
language into text.

vi. Answer: Cross-validation 1

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

ii.
Answer: Confusion Matrix 1

iii. (b) Histogram 1

iv. (b) Speech Recognition 1

v. (b) Natural Language Processing (NLP) 1

vi. (c) Medical diagnosis 1

SECTION B: SUBJECTIVE TYPE QUESTIONS

Mark
s
Q.
QUESTION
No.

Q. 6 Answer:Feedback is the receiver's response to a message, which helps the sender know 2
whether the message was understood correctly. It is important because it ensures effective
communication by allowing for adjustments if the message was not clearly understood.

Q. 7 Answer: Setting SMART goals—Specific, Measurable, Achievable, Relevant, and Time- 2

bound—helps individuals stay focused, organize their time effectively, and achieve their
objectives efficiently. It provides a clear roadmap and motivates one to stay on track.

Q. 8 Answer: The cause may be malware or adware infecting her computer. Aisha should run an
antivirus scan to detect and remove any malicious software to restore her computer's performance.

Q. 9 Answer: Key characteristics of an entrepreneur include creativity, risk-taking, resilience, and 2

leadership, which enable them to identify opportunities and navigate challenges in business

417 - ARTIFICIAL INTELLIGENCE – X PAGE 3 OF 8

Q 10 . Answer: Renewable energy sources, such as solar, wind, and hydroelectric power, provide 2
sustainable alternatives to fossil fuels. They help reduce greenhouse gas emissions, decrease
reliance on finite resources, and contribute to energy security.

Answer any 4 out of the given 6 questions in 20 – 30 words each (2 x 4 = 8

marks)
Q. 11 Answer: AI chatbots provide 24/7 customer support, reduce response time for queries, and free up (2 marks
human agents to handle more complex issues, improving overall service efficiency. for any
correct
explanati
on)
Q. 12 Answer: Model evaluation is essential to determine how well the AI model performs on (2 marks
unseen data, ensuring its effectiveness in real-world scenarios. Metrics they could use include for any
accuracy, precision, recall, and F1- score. correct
explanati
on

Q. 13 Answer: Supervised learning involves training a model on labeled data, where the input-output 2 marks
pairs are known, to make predictions. In contrast, unsupervised learning deals with unlabeled for any
data, where the model identifies patterns and structures without predefined labels. correct
explanati
on

Q. 14  Facial Recognition: Used in security systems, smartphones, and social media platforms ( 1marks
to identify and verify individuals based on their facial features. each for
CORREC
 Autonomous Vehicles: Enables self-driving cars to detect obstacles, recognize traffic T ANY
signs, and navigate roads safely. TWO
APPLIC
 Medical Imaging: Assists in analyzing medical images (such as X-rays, MRIs, and CT ATION )
scans) to detect diseases and anomalies, improving diagnosis and treatment.

 Object Detection: Used in retail for inventory management, where cameras can identify
products on shelves and track stock levels.

 Surveillance Systems: Monitors public and private spaces for security purposes,
detecting unusual activities or identifying individuals in real-time.

 Image Search Engines: Allows users to search for similar images by uploading a photo,
which the system analyzes to find visually similar content online.

 Gesture Recognition: Used in gaming and interactive applications to interpret user

movements and gestures, enhancing user engagement.

417 - ARTIFICIAL INTELLIGENCE – X PAGE 4 OF 8

Q. 15 Answer: Stop words are common words (like "is," "the," "and") that carry little (2 marks
meaning in analysis and are often removed from text data to reduce noise and improve for any
the efficiency and accuracy of NLP algorithms. correct
explanati
on)

Q. 16 2

Predicted Spam Predicted Not Spam

CONFUSION MATRIX
Actual Spam 40 (TP) 20 (FN)
Actual Not Spam 10 (FP) 30 (TN)

Answer any 3 out of the given 5 questions in 50– 80 words each (4 x 3 = 12

marks)
Q. 17 Answer: 2+2

 a) Recommendation systems utilize machine learning algorithms to analyze user data,

such as browsing history and purchase behavior, to identify patterns and preferences.
They then generate personalized product suggestions that enhance the shopping
experience.
 b) Two advantages include:
o For consumers, personalized recommendations help discover products they
might not have found otherwise, improving their shopping experience.
o For businesses, AI-driven recommendations can increase sales and customer
retention by providing targeted marketing and enhancing customer
satisfaction.

Q. 18 4
Answer: Data preprocessing is a crucial step in the AI project cycle because it prepares the
raw data for analysis, ensuring that it is clean, consistent, and usable. Poorly prepared data
can lead to inaccurate models and unreliable results. Common techniques used in data
preprocessing include:

 Data Cleaning: Removing or correcting inaccuracies, duplicate records, and missing

values to improve data quality.
 Normalization: Scaling data to a standard range (e.g., 0 to 1) to ensure that no single
feature disproportionately influences the model.
 Encoding Categorical Variables: Converting categorical data into numerical format
using techniques like one-hot encoding, making it suitable for algorithm processing.
 Feature Selection: Identifying and selecting the most relevant features from the
dataset that contribute to the model’s predictive power.

Effective data preprocessing enhances model performance and increases the likelihood of
achieving accurate results.

Q. 19 Answer: Data collection is a foundational step in the AI project cycle that involves 4
417 - ARTIFICIAL INTELLIGENCE – X PAGE 5 OF 8
 gathering the necessary data to train and evaluate AI models. The quality and quantity of
the data directly impact the model's performance and its ability to make accurate
predictions.

Common sources of data for AI projects include:

 Public Datasets: Pre-existing datasets available from government organizations,

research institutions, or online repositories (e.g., Kaggle,
UCI Machine Learning Repository).
 Surveys and Questionnaires: Custom data can be collected through surveys or
questionnaires targeted at specific audiences to gather
insights.
 Web Scraping: Extracting data from websites using automated tools to gather
large volumes of information.

IoT Devices: Data collected from sensors and devices that provide real-time information,
particularly in fields like healthcare and smart cities.

Q 20 To normalize this feedback, follow these steps: 4

mar
1. Convert to lowercase: Convert all text to lowercase to make analysis uniform. ks
o Result: "the service was excellent!!! but the delivery was late... very slow
response :("
2. Remove punctuation: Eliminate unnecessary punctuation.
o Result: "the service was excellent but the delivery was late very slow
response"
3. Correct spelling (Optional): You may want to correct spelling errors for further
clarity, especially if feedback contains typos.
o Result: If there were misspelled words like "excellnt" or "responce," they
would be corrected to "excellent" and "response."
4. Remove stop words (Optional): For specific tasks, words like "the" and "was" could
be removed if they don’t add value to analysis.
o Result (if stop words are removed): "service excellent delivery late slow
response"

The normalized feedback is: "service excellent delivery late slow response"

Q 21 Ans: (i)the total number of wrong predictions made by the model is the sum of false positive and 4
false negative.

FP+FN=40+12

= 52

(ii) Precision=TP/(TP+FP)

=50/(50+40) 50/90

=0.55
417 - ARTIFICIAL INTELLIGENCE – X PAGE 6 OF 8
 Recall=TP/(TP+FN) =50/(50+12) =50/62 =.81

F1 Score = 2*Precision*Recall/(Precision+Recall)

=2*0.55*.81/(.55+.81)

=.891/1.36

=0.65

(1 marks for part (i) and ½ mark for each formula and ½ mark each for substitution of values in
part(ii)) Please note: the mathematical calculations can be ignored

************************************************************************************************

417 - ARTIFICIAL INTELLIGENCE – X PAGE 7 OF 8