You are on page 1of 5

ARTIFICIAL INTELLIGENCE (SUBJECT CODE - 417)

Sample Question Paper 2


Answer Key
Class X
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.
ps
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.
Ki
7. SECTION B- SUBJECTIVE TYPE QUESTIONS (26 MARKS):
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.

Section A (Employability Skills)


Answer any 3 questions out of the given 4 (one mark each)

1. “Decision-making is the process of making choices by identifying a situation, collecting


relevant information and evaluating alternatives.”
Is the above statement correct? Yes.
2. Briefly explain any one element of the Big Five Personality model.
Openness is one of the elements of the Big Five model. It refers to the ability of an individual
to accept new things.

© Kips Learning Pvt. Ltd 2023


3. The Sustainable Development Goals (SDGs) were formulated for addressing challenges
related to three broad areas or factors. List out these three.
The three factors or broad areas related to SDGs are environment, social and economy.
4. Define ethics.
Ethics can be defined as the science of moral duty and ideal human behaviour.

Answer any 1 question out of the given 2 questions (2 marks each)

5. How can achieving Sustainable Development Goals (SDGs) improve the position of women?
SDG 5 (gender equality and empowerment of women and girls) is directly related to
improving position of women. Many other SDGs indirectly work towards women
empowerment. For example, SDG 3 with healthy lives and SDG 4 with inclusive and
equitable education. Achievement of these will make lives of females and women better.
6. Differentiate between structured and un-structured decisions.
Structured decisions are decisions which have an already known solution. They do not
require any decision support. Un-structured decisions do not have any agreed upon solution.
Their solutions depend on the choice of the decision maker.

ps
Section B (Subject Specific Skills)
Answer any 4 questions out of the given 6 questions (one mark each)

7. What is accuracy?
Accuracy is a measure of the trueness of the predictions of the model. It provides
percentage of correct predictions out of the observations made.
8. What is the main function of pool layer in the Convolutional Neural Networks?
The main function of the pool layer in the Convolutional Neural Network is to crop the size
of the volume of the data.
9. What is real time data?
Ki
Real time data is the data which is available in real time i.e., is available as soon as the
related event takes place.
10. “Gaming systems like Microsoft Kinect use Smart Interaction technology”
Is this statement correct? Yes
11. List any two practical uses of Natural Language Processing.
Natural Language Processing is used for language translation applications for example
Google Translate. It is also used in digital assistants like Ok Google and Siri.
12. What is OpenCV?
OpenCV is an open source computer vision and machine learning software library. OpenCV
was created to offer a shared infrastructure for CV applications.

Answer any 4 questions out of the given six questions (two marks each)

13. What is sentiment analysis? Why is it used?


Sentiment analysis is used for finding sentiment in the text. Sentiment analysis is used for
identifying opinions and the feelings of the customers. For example, the customer’s
experience with a particular product or service or understanding the purchasing decisions
taken by the customer.
14. Briefly explain deep learning.

© Kips Learning Pvt. Ltd 2023


Deep learning is a subunit (in certain sense also an extension) of machine learning. It is a
technique used for preparing Artificial Intelligence for application. It makes use of artificial
neural network for decision making and solution predictions.
15. Differentiate between discrete and continuous numerical data.
Discrete numerical data is made up of whole numbers. It cannot contain decimal values. For
examples number of people in a family. Continuous numerical data on the other hand can
take any value within a range. For example, the weight of vegetables.
16. What is algorithm?
Algorithms are one of the essential components of artificial intelligence systems. They are
mathematical formulas and programming commands which allow non-intelligent computers
to solve problems using artificial intelligence.
17. Flip the following mask correctly for performing the convolution of an image.
11 21 5
2 10 9
13 5 7

For performing convolution of an image, we have to first flip the mask horizontally and then
vertically. Flipping the mask horizontally will provide the following matrix.
ps
5 21 11
9 10 2
7 5 13
Flipping it again vertically will provide the following matrix.

7 5 13
9 10 2
5 21 11

18. Lemmatization is an important aspect of natural language processing. Fill the following table
with the correct lemma for the given word forms.
Ki
Form Lemma
Amused Amuse
Floated Float
Boundaries Boundary
Using Use

Section C (Competency Based Questions)


Answer any 2 questions out of the given 3 questions (4 marks each)

19. Differentiate between the supervised learning and unsupervised learning with an example.
In supervised learning, a labelled dataset is used for training the AI model. For example, a
training a model with the dataset containing different images of cats. After training, the
model will be able to process the input image and tell if it is a cat or not.
In unsupervised learning, the dataset used for training the AI model are neither classified nor
labeled. The model uses the training dataset for finding hidden patterns in the data. For
example, an AI model trained on such datasets will be able to club together images of

© Kips Learning Pvt. Ltd 2023


different animals like cats and dogs. However, it will not be able to tell if a particular image is
that of a cat.

20. Increased frequency of rain in a city leads to traffic jams and frequent school closures. ABC
school has created an AI model for improving rain management and providing timely school
closure information to the stakeholders. The confusion matrix for the model is given below:
Actuals
Positive Negative
Positive 13 5
Predictions
Negative 7 14
Explain the process of calculating the F1 score based on the above data.
F1 score is calculated based on precision and recall.
Step 1: Calculate the precision of the model
The precision can be calculated in the following manner:
True Positives
Precision =
All Predicted Positives
13
Precision = = 0.7222
13 + 5
Step 2: Calculate the recall of the model.
This can be done with the help of the following formula
ps True Positives
Recall =
True Positives + False Negatives
13
Recall = = 0.65
13 + 7
Step 3: Calculate the F1 Score
This can be done with the help of the following formula
Precision × Recall
F1 Score = 2( )
Precision + Recall
0.7222 × 0.65
F1 Score = 2 � � = 68.42
0.7222 + 0.65
Ki
21. Consider the following data:
Document 1: I like to take long walks
Document 2: Rajesh likes to swim daily
Document 3: Kavita loves singing

Apply the four steps of the Bag of Words model on the above data and generate the output.
Machine learning algorithms cannot use raw text. The text first needs to be converted into
vectors of numbers. Bag of words is one of the methods of this conversion.

Step 1 of Bag of Words Model: Text Normalization.


Text normalization requires conversion of the text to a lower level. This is done with the help
of sentence and work tokenization and text lemmatization and stemming. There is no need
for sentence tokenization in the given data. Word tokenization simply means dividing each
sentence into words. The result will be as follows:
Document 1 (I, like, to, take, long, walks)
Document 2 (Rajesh, likes, to, swim, daily)
Document 3 (Kavita, loves, singing)

© Kips Learning Pvt. Ltd 2023


Text stemming and lemmatization is done for obtaining the base or dictionary form of
words. Stemming is done by chopping off the ends of the words. Lemmatization is done with
the help of vocabulary and morphological analysis. The result will be as follows:
Document 1 (I, like, to, take, long, walk)
Document 2 (Rajesh, like, to, swim, daily)
Document 3 (Kavita, love, sing)

After stemming and lemmatization the stopwords, special characters and numbers are
removed from the text. The result will be as follows:
Document 1 (I, like, take, long, walk)
Document 2 (Rajesh, like, swim, daily)
Document 3 (Kavita, love, sing)

Then all the text is converted to a common case. The result will be as follows:
Document 1 (i, like, take, long, walk)
Document 2 (rajesh, like, swim, daily)
Document 3 (kavita, love, sing)

Step 2: Creating Dictionary


ps
For creating dictionary, first all the unique words are listed.
i like take long walk rajesh swim daily kavita love sing

Step 3: Creating document vector

Now the documents are checked for the unique words. Value of 1 is given for the first
appearance of the unique word. This is incremented by 1 for every additional presence.
Value of 0 is given if the word is absent from the document. The result will be as under.

i like take long walk rajesh swim daily kavita love sing
Document 1 1 1 1 1 1 0 0 0 0 0 0
Ki
Step 4: Create document vector for all the documents

i like take long walk rajesh swim daily kavita love sing
Document 1 1 1 1 1 1 0 0 0 0 0 0
Document 2 0 1 0 0 0 0 1 1 0 0 0
Document 3 0 0 0 0 0 0 0 0 1 1 1

© Kips Learning Pvt. Ltd 2023

You might also like