CBSE | DEPARTMENT OF SKILL EDUCATION

ARTIFICIAL INTELLIGENCE (SUBJECT CODE - 843)

MARKING SCHEME FOR CLASS XII (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.No. Unit/ Marks
QUESTION Chap.
No.
Q. 1
i. c. To whom? 1 1
Explanation: An indirect object answers the question "to whom" or "for whom" an
action is done.

ii. a. Intrinsic motivation 2 1

Explanation: The person is motivated by internal satisfaction and enjoyment in
learning, which is intrinsic motivation.

iii. b. Statement 1 is correct but Statement 2 is incorrect 2 1

Explanation: Empathy is important in both personal and professional settings,
making Statement 2 incorrect.

iv. a. A - 3; B - 2; C - 1 3 1

v. d. Empathy 4 1
Explanation: Rina is showing an understanding of her colleagues' struggles and is
offering help, which reflects empathy.

vi. c. Use a certified e-waste recycler or disposal partner 5 1

Explanation: Certified recyclers handle e-waste properly, recovering valuable
materials and reducing environmental pollution from harmful substances.

843 – ARTIFICIAL INTELLIGENCE – Page 1 of 5

XII
Q. 2
i. d. Ideate 1 1
Explanation: The Ideate phase in Design Thinking focuses on brainstorming and
exploring diverse ideas for potential solutions.

ii. b. Understanding the audience 3 1

Explanation: Understanding the audience helps shape the data story to be relevant
and impactful.

iii. a. 1=B, 2=A, 3=C, 4=D 1 1

Explanation: Classification identifies spam emails, regression predicts continuous
values (like house prices), clustering groups customers, and anomaly detection
identifies unusual patterns like fraud.

iv. d. 1 & 4 2 1

v. c. iterative 1 1
Explanation: Iterative processes involve repeated cycles of testing and refining to
improve model accuracy.

vi. a. Permanent 2 1
Explanation: Deployment is generally a permanent phase, as it involves putting the
model into production, making it accessible to users, and ensuring it serves its
intended purpose.

Q. 3
i. b. Data Preparation 1 1
Explanation: Data Preparation is the stage where data is organized and cleaned
to ensure it is ready for analysis.

ii. a. Engaging 3 1
Explanation: Engaging stories capture the audience's attention by connecting
with them on a personal level.

iii. c. Design 2 1
Explanation: The Design phase is where models are created, and algorithms are
selected based on project goals.

iv. d. A is False but R is true 3 1

v. c. Lower RMSE values indicate better model performance. 1 1

Explanation: RMSE measures the model's error rate; a lower value indicates
less error and, hence, better performance.

vi. b. deployment 2 1

Q. 4
i. a. Scoping 2 1
Explanation: Scoping is the initial phase where goals, success metrics, and
project expectations are defined.

ii. a. Narrative b. Data 3 1

843 – ARTIFICIAL INTELLIGENCE – Page 2 of 5

XII
 iii. c. It requires multiple training iterations, making it computationally demanding. 1 1
Explanation: Cross-validation requires re-training the model several times,
which can be computationally intensive.

iv. b. Improve training efficiency and model accuracy 3 1

Explanation: Proper tuning of hyperparameters ensures that the model trains
effectively and achieves higher accuracy.

v. c. 1-3-2-4 2 1
Explanation: The process starts by organizing data, then identifying
relationships, visualizing, and finally crafting a narrative with insights.

vi. d. Regression predicts quantities, and classification predicts labels. 1 1

Explanation: Regression outputs a continuous variable, while classification
outputs a discrete label or category.

Q. 5
i. d. numbers 3 1
Explanation: Stories can make numbers or statistical data more engaging and
relatable.

ii. b. Medical diagnosis 1 1

Explanation: AI models are commonly used to predict outcomes in medical
diagnosis.

iii. d. (1), (2), and (4) 2 1

Explanation: Performance testing, model interpretation, and validation are
essential to ensuring model accuracy and reliability.

iv. c. Weather forecast 1 1

v. a. less than 100 2 1

Explanation: A lower Mean Squared Error (MSE) typically indicates a more
accurate model.

vi. b. structured patterns 1 1

SECTION B: SUBJECTIVE TYPE QUESTIONS

Q.No. QUESTION Unit/ Marks
Chap. No.
Q. 6 "E" stands for Empathy. 1 2
Empathy is essential in active listening as it involves understanding and sharing
the feelings of the speaker. By being empathetic, listeners can connect with the
speaker on a deeper level, fostering a supportive and respectful conversation.
(1 mark for the word ‘Empathy ’; and 1 mark for explanation)

Q. 7 Anita can try exposure and response prevention (ERP) therapy. 2 2

 This technique gradually exposes her to situations that trigger her anxiety
 while helping her learn to resist the urge to rearrange her belongings,
 ultimately reducing her compulsive behavior.
(Any 2 of the above – 1 mark each)
843 – ARTIFICIAL INTELLIGENCE – Page 3 of 5
XII
Q. 8 A workbook helps 3 2
 manage multiple related worksheets under a single file,
 allowing users to organize and access various datasets conveniently.
 This enables better project management and data analysis across different
sheets.
(Any 2 of the above – 1 mark each)

Q. 9 The Make in India initiative supports small-scale industries by 4 2

 promoting manufacturing and entrepreneurship,
 providing various incentives, and
 improving ease of doing business,
 which encourages investment and job creation.
(Any 2 ; 1 mark each)

Q. 10 Green jobs promote energy efficiency by 5 2

 developing and implementing technologies that reduce energy
consumption,
 which lowers costs for businesses and
 households while decreasing environmental impact.
(Any 2 ; 1 mark each)

Q. 11 A capstone project differs from traditional assignments in that it is usually more 1 2

comprehensive and interdisciplinary, requiring students to conduct independent
research, collaborate with peers, and demonstrate critical thinking, problem-
solving, and communication skills in a practical context.

Q. 12 It is impossible for MSE to take on negative values because the squaring of 1 2

differences between predicted and actual values results in non-negative numbers.
The average of these squared values will also be non-negative, making MSE either
zero (in the case of perfect predictions) or positive.

Q. 13 A strong narrative enhances character development by providing context for the 3 2

characters' actions, motivations, and growth throughout the story. It allows
readers to connect with characters on an emotional level, understanding their
challenges and transformations.

Q. 14 Considerations should include: 1 2

 What essential features must the app include for its core functionality?
 How will user experience (UX) design affect the app's usability?
 What additional features could enhance user engagement and retention?

Q. 15 It is important to have a separate test set to evaluate the model's performance on 1 2

unseen data. This helps to assess the model's generalization ability and ensures
that it is not merely memorizing the training data.

Q. 16 The Build phase involves the actual implementation of the model as defined in the 2 2
Design phase. This includes coding the algorithms, training the model on the
training dataset, and iterating on the design to improve performance, whereas the
Design phase focuses on planning and strategy.

Q. 17 Effective data storytelling can influence decision-making by providing clarity, 3 4

guiding strategic discussions, and facilitating understanding among stakeholders.

843 – ARTIFICIAL INTELLIGENCE – Page 4 of 5

XII
 Diagram Explanation:
The diagram could depict a flow from data collection → data analysis → data
visualization → narrative creation → informed decision-making, emphasizing how
each step builds upon the previous to drive action.

Q. 18 RMSE values indicate the average distance between the predicted and actual 1 4
values. A lower RMSE value signifies a better fit of the model to the data, while a
higher RMSE indicates larger errors. In context, RMSE can be used to compare the
performance of different models, helping to select the most accurate one.

Q. 19 Using a balanced test dataset is important to prevent bias in performance 2 4

evaluation. An imbalanced dataset can skew results, making a model appear more
accurate than it is, particularly in classification tasks where one class may
dominate.

Q. 20  In k-fold cross-validation, the dataset is divided into k equally sized subsets (or 1 4
folds).
 The model is trained on k-1 folds and validated on the remaining fold.
 This process is repeated k times, with each fold serving as the validation set
once.
 The final performance metric is the average of the k individual tests, providing
a robust evaluation of the model.

Q. 21 Identifying the target audience is significant because 3 4

 it influences the narrative style, data presentation, and
 overall message of the story.
 Understanding the audience’s background and interests allows the storyteller
to tailor the content to resonate with them,
 enhancing impact and engagement.

843 – ARTIFICIAL INTELLIGENCE – Page 5 of 5

XII