CBSE Class 12 Artificial Intelligence 843 Practice Paper Set 1

CBSE Class 12 Artificial Intelligence 843 Practice Paper Set 1

SAMPLE QUESTION PAPER (SET-1) [2025-26]

ARTIFICIAL INTELLIGENCE (SUBJECT CODE – 843)

CLASS XII

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 carrying 1 mark each.
4. Section B contains Subjective type questions carrying 2 marks and 4 marks each.
5. Out of the given (5 + 16 =) 21 questions, a candidate has to answer 15 questions.
6. All questions of a particular section must be attempted in the correct order.

SECTION A: OBJECTIVE TYPE QUESTIONS (24 MARKS)

Q. 1 Answer any 4 out of the 6 questions given on Employability Skills (1 mark each):

1. Which of the following is considered a “Barrier to Communication” arising from linguistic differences?

(a) Noise                    (b) Jargon                          (c) Lack of Interest      (d) Physical Distance

2. Rohan has a deadline to meet but spends hours organizing his desk and checking emails instead of starting the actual work. This behavior is known as:

(a) Prioritization    (b) Procrastination     (c) Delegation                (d) Scheduling

3. Which personality disorder is characterized by a pattern of social inhibition, feelings of inadequacy, and hypersensitivity to negative evaluation?

(a) Avoidant Personality Disorder             (b) Borderline Personality Disorder
(c) Narcissistic Personality Disorder       (d) Antisocial Personality Disorder

4. Which function in a spreadsheet is used to find the total of two numbers.

(a) sum()                   (b) total()                           (c) add()                             (d) sums()

5. The ability of an entrepreneur to bounce back from failures and keep moving forward is called:

(a) Innovation         (b) Resilience                 (c) Flexibility                    (d) Independence

6. Which of the following is a non-renewable energy source?

(a) Wind Energy     (b) Solar Energy            (c) Natural Gas             (d) Biomass

Q. 2 Answer any 5 out of the given 6 questions (1 mark each):

1. Which Python library is commonly used for visualizing data in graphs like histograms and scatter plots?

(a) Pandas               (b) Matplotlib                 (c) NumPy                        (d) Scikit-learn

2. In a Convolutional Neural Network (CNN), what is the primary purpose of the “Kernel” or “Filter”?

(a) To reduce the image size.                       (b) To extract features like edges and textures.

(c) To classify the final output.                    (d) To flatten the image into a vector.

3. The hidden insights and useful patterns extracted from Big Data that help in business decision-making refer to which “V”?

(a) Volume               (b) Value            (c) Velocity                       (d) Variety

4. A neural network with multiple hidden layers is specifically referred to as:

(a) Deep Learning Network            (b) Shallow Network
(c) Perceptron                                        (d) Linear Regressor

5. Which of the following is NOT a capability of current Generative AI models?

(a) Writing Code                                            (b) Composing Music
(c) Sentience (Consciousness)                 (d) Creating 3D Models

6. When choosing colors for a data visualization, you should avoid:

(a) Using high-contrast colors for key data.
(b) Using too many similar shades that are hard to distinguish.
(c) Using a legend to explain colors.
(d) Matching colors to the brand theme.

Q. 3 Answer any 5 out of the given 6 questions (1 mark each):

1. In the “Data Understanding” phase, exploring the data to discover initial insights or quality issues is often called:

(a) EDA (Exploratory Data Analysis)        (b) ETL (Extract, Transform, Load)
(c) API Integration                                                (d) Model Training

2. The formula (TP + TN) / (TP + TN + FP + FN) is used to calculate:

(a) Precision            (b) Recall                          (c) Accuracy                    (d) F1 Score

3. Which computer vision application allows a car to stay in its lane automatically?

(a) Face Recognition                                               (b) Lane Detection
(c) Optical Character Recognition (OCR)            (d) Image Captioning

4. Emails, social media posts, and customer reviews are examples of:

(a) Structured Data                                            (b) Semi-structured Data
(c) Unstructured Data                                      (d) Metadata

5. In a neural network, the “Bias” value is added to the weighted sum to:

(a) Make the calculation faster.                 (b) Shift the activation function curve.
(c) Reduce the number of inputs.             (d) Eliminate the need for weights.

6. In Freytag’s Pyramid, the moment of greatest tension or the “major insight” of the story is known as the:

(a) Exposition                 (b) Rising Action           (c) Climax         (d) Resolution

Q. 4 Answer any 5 out of the 6 questions given (1 mark each):

1. Assertion (A): We split data into Training and Testing sets.

Reason (R): Using the same data for training and testing causes “Data Leakage” and gives a false sense of high accuracy.

(a) Both A and R are true, and R is the correct explanation of A.
(b) Both A and R are true, but R is NOT the correct explanation of A.
(c) A is true, but R is false.
(d) A is false, but R is true.

2. In a Convolutional Neural Network (CNN), which layer is primarily responsible for automatically detecting features like lines, edges, and textures?

(a) Fully Connected Layer                                      (b) Convolutional Layer
(c) Flattening Layer                                                 (d) Output Layer

3. Which technology is primarily used to distribute the processing of Big Data across clusters of computers?

(a) Hadoop/MapReduce                                         (b) Microsoft Excel
(c) Bluetooth                                                             (d) Adobe Photoshop

4. Which of the following is a “Supervised Learning” task?

(a) Clustering customers based on buying behavior.
(b) Predicting the price of a house based on historical data.
(c) Reducing the dimensions of a dataset.
(d) Finding associations between products.

5. The “Input Layer” of a neural network:

(a) Performs complex calculations.
(b) Passes the raw data to the hidden layers.
(c) Generates the final prediction.
(d) Calculates the error/loss.

6. Deepfakes are created using which specific AI architecture?

(a) Decision Trees                       
(b) K-Nearest Neighbors
(c) GANs (Generative Adversarial Networks)
(d) Linear Regression

Q. 5 Answer any 5 out of the given 6 questions (1 mark each):

1. A “False Positive” (Type I Error) occurs when:

(a) The model predicts positive, and the actual value is positive.
(b) The model predicts positive, but the actual value is negative.
(c) The model predicts negative, and the actual value is negative.
(d) The model predicts negative, but the actual value is positive.

2. Assertion (A): RGB images have 3 channels (Red, Green, Blue).

Reason (R): Grayscale images contain more information than RGB images.

(a) Both A and R are true, and R is the correct explanation of A.
(b) Both A and R are true, but R is NOT the correct explanation of A.
(c) A is true, but R is false.
(d) A is false, but R is true.

3. A pixel value of 0 in a standard 8-bit grayscale image represents:

(a) White                            (b) Grey                              (c) Black                            (d) Transparent

4. When a model learns the details and noise in the training data to the extent that it negatively impacts the performance of the model on new data, it is called:

(a) Underfitting              (b) Overfitting                 (c) Regularization        (d) Optimization

5. The text input given to an LLM to guide its output is technically called a:

(a) Command                (b) Script                           (c) Prompt                        (d) Code

6. Which critical step in Data Storytelling involves understanding the viewers’ technical knowledge to decide the complexity of the presentation?

(a) Data Cleaning                                   (b) Knowing your Audience
(c) Choosing the Color Palette           (d) Data Collection

SECTION B: SUBJECTIVE TYPE QUESTIONS (26 MARKS)

Answer any 3 out of the given 5 questions on Employability Skills (2 marks each):

Answer any 4 out of the given 6 questions in 20 – 30 words each (2 marks each):

Answer any 3 out of the given 5 questions in 50– 80 words each (4 marks each):

Answer Key

SECTION A: OBJECTIVE TYPE QUESTIONS

Q. 1 (Employability Skills)

1. (b) Jargon
2. (b) Procrastination
3. (a) Avoidant Personality Disorder
4. (a) SUM( )
5. (b) Resilience
6. (c) Natural Gas

Q. 2 (Subject Specific Skills)

1. (b) Matplotlib
2. (b) To extract features like edges and textures.
3. (b) Value
4. (a) Deep Learning Network
5. (c) Sentience (Consciousness)
6. (b) Using too many similar shades that are hard to distinguish.

Q. 3 (Subject Specific Skills)

1. (a) EDA (Exploratory Data Analysis)
2. (c) Accuracy
3. (b) Lane Detection
4. (c) Unstructured Data
5.  (b) Shift the activation function curve.
6. (c) Climax

Q. 4 (Subject Specific Skills)

1. (a) Both A and R are true, and R is the correct explanation of A.
2. (b) Convolutional Layer
3. (a) Hadoop
4. (b) Predicting the price of a house based on historical data.
5. (b) Passes the raw data to the hidden layers.
6. (c) GANs (Generative Adversarial Networks)

Q. 5 (Subject Specific Skills)

1. (b) The model predicts positive, but the actual value is negative.
2. (c) A is true, but R is false.
3. (c) Black (0 is black, 255 is white).
4. (b) Overfitting
5. (c) Prompt
6. (b) Knowing your Audience