CONTINUOUS ASSESSMENT TEST – 2
Regulations R 2019 - V21
Department of Computer Science and Engineering
Final Year / Seventh Semester
191CSC701T – DATA SCIENCE
CO1: Develop a big data solution using Hadoop and data visualization technique.
CO2: Use Exploratory Data Analysis and statistical inference for real world applications.
CO3: Analyze the data using regression and Matrix decomposition techniques.
CO4: Choose appropriate pre-processing and feature selection techniques to lever
inconsistencies.
CO5: Apply Basic Machine Learning Algorithms for complex problems.
UNIT III - REGRESSION MODELS & LINEAR ALGEBRA BASICS
CO’s Bloom’s
Q.No Questions
Level
PART A
What is the purpose of Singular Value Decomposition (SVD) in
1. CO3 K1
matrix decomposition?
2. Define partial correlation. CO3 K1
Differentiate between multiple correlation and partial correlation in CO3
3. K1
the context of correlation analysis.
4. Describe the relationship between PCA and matrix decomposition CO3 K2
5. Define multiple correlation CO3 K2
6. What is Dimensionality Reduction? CO3 K2
7. Define Principal Component Analysis CO3 K1
8. List out the applications of singular Value Decomposition CO3 K1
What type of matrix can be used to represent relationships between CO3
9. K2
variables in a dataset?
What are the three matrices produced by Singular Value CO3
10. K1
Decomposition (SVD)?
11. What is the role of PCA in data analysis? CO3 K2
12. Define regression. CO3 K1
13. When do we apply the clustering method in real time? CO3 K1
14. Define covariance matrix. CO3 K2
15. List out the applications of PCA CO3 K1
CO’s Bloom’s
Q.No Questions
Level
Part – B
1. Explain in detail about Linear algebraic operations on matrices. CO3 K2
Describe how principal component analysis is carried out to reduce
2. CO3 K2
dimensionality of data sets.
Explain the concept of multiple correlation. How is the multiple
3. correlation coefficient calculated? Discuss its significance in CO3 K2
regression analysis.
What is partial correlation? Explain its importance and how it differs
4. from simple correlation. How do you calculate partial correlation CO3 K2
between two variables while controlling for a third variable?
Discuss key linear algebraic operations on matrices such as matrix
5. addition, multiplication, and inversion. How are these operations CO3 K2
used in data analysis?
What is Singular Value Decomposition (SVD)? Explain the steps
6. CO3 K2
involved in SVD and its significance in data analysis
Explain how matrices are used to represent relationships between
7 data. Provide an example of using a matrix to model a real-world CO3 K2
relationship in data.
Explain in detail about the relationship between PCA(Principal
8 CO3 K2
Component Analysis) and matrix decomposition.
CO’s Bloom’s
Q.No Questions
Level
Part c
Given data = {2, 3, 4, 5, 6, 7; 1, 5, 3, 6, 7, 8}. Compute the principal
1. component using PCA Algorithm. CO3 K3
Discuss key linear algebraic operations on matrices such as matrix
2. addition, multiplication, and inversion. How are these operations used CO3 K3
in data analysis?
UNIT – IV DATA PREPROCESSING AND FEATURE SELECTION
Q.N CO’s Bloom’s
Questions
o Level
Part a
Explain the importance of data cleaning in the context of machine
1. CO4 K1
learning and data analysis.
2. Define normalization. CO4 K1
3. Explain the concept of data reduction and its benefits in data analysis CO4 K2
Compare data transformation and data discretization techniques in the
4. CO4 K1
context of preparing data for machine learning.
Compare filter and wrapper methods for feature selection in machine
5. CO4 K2
learning.
6. What do you mean by Data Discretization? CO4 K2
7. Explain about data reduction. CO4 K2
8. Define data Compression CO4 K1
9. List out the Feature Selection algorithms CO4 K1
Discuss the importance of feature generation and feature selection in
10. CO4 K2
machine learning.
11. List out the methods used to handle the missing values. CO4 K2
12. Define binning and its types. CO4 K2
13. Define feature Engineering. CO4 K2
14. What do you mean by over fitting? CO4 K2
15. Define Noisy data. How will you eliminate noisy data? CO4 K2
Q. CO’s Bloom’s
Questions
No Level
Part b
Explain the importance of data cleaning in the data pre-processing
1. CO4 K2
phase. Discuss common techniques used for data cleaning
Describe a case study where data cleaning played a pivotal role in
2. improving data analysis outcomes. Outline the steps involved in the data CO4 K3
cleaning process for this case study
Explain the concept of data reduction. Discuss different techniques used
3. CO4 K2
for dimensionality reduction in large datasets
Define data transformation and discuss its role in preparing data for
4. CO4 K3
analysis. Provide examples of common data transformation techniques
Discuss the advantages and limitations of Filter methods in feature
5. selection. Provide examples of commonly used Filter methods and their CO4 K3
applications.
Define data integration and discuss its significance in combining
6. CO4 K3
heterogeneous data sources
Explain the concept of data discretization. How does it aid in handling
7 CO4 K3
continuous data in machine learning?
8 Explain in detail about detail about Feature selection algorithms CO4 K3
Q. CO’s Bloom’s
Questions
No Level
Part- C
Discuss the importance of feature selection in machine learning.
1. CO4 K4
Compare and contrast Filter and Wrapper methods for feature selection.
Illustrate a scenario where data reduction techniques were applied to
2. improve computational efficiency and model performance. Outline the CO4 K5
steps involved in data reduction for this scenario.
UNIT – V BASIC MACHINE LEARNING ALGORITHMS
Q. CO’s Bloom’s
Questions
No Level
Part a
1. List the advantages and disadvantages of decision tree. CO5 K1
2. Define information gain. CO5 K1
3. What is SVM? CO5 K1
4. What is the goal of ensemble method? CO5 K1
5. What are the two families of ensemble methods? CO5 K1
6. What is reinforcement learning? CO5 K1
What is the difference between artificial intelligence and machine
7. CO5 K1
learning methods?
8. What is supervised learning? CO5 K1
9. What is bagging? CO5 K2
10. List out the applications of Association rules. CO5 K2
11. What is the Purpose of Apriori Algorithm? CO5 K1
12. List out the applications of Association rules. CO5 K1
13. Define the concept of classification CO5 K1
14. How will you solve a classification problem using Decision Tree? CO5 K1
15. Define k-means clustering. CO5 K2
Q. CO’s Bloom’s
Questions
No Level
Part – B
1. Explain in detail about the k-nearest neighbour algorithm with example. CO5 K3
2. Explain the concept of random forests algorithm in detail CO5 K3
Use decision tree to classify the students in a class based on their
3. CO5 K3
academic performance.
4. Explain in detail about Naïve Bayes Classification. CO5 K3
5. How evaluation is performed on decision trees? CO5 K3
6. Explain the significance of ensemble methods. CO5 K3
7 Explain association rules in detail. CO5 K3
8 Explain in detail about SVM (Support Vector Machines) CO5 K3
Q. CO’s Bloom’s
Questions
No Level
Part c
Construct a decision tree to determine whether the animal lays eggs
with the ID3 Algorithm.
Animal Warm blooded Feathers Fur Swins Lays eggs
Ostrich Yes Yes No No Yes
Crocodile No No No Yes Yes
1 CO5 K3
Raven Yes Yes No No Yes
Albatross Yes Yes No No Yes
Dolphin Yes No No Yes No
Koala Yes No Yes No No
Summarize K-means algorithm and group the points (1, 0, 1), (1, 1, 0),
2. CO5 K3
(0, 0, 1) and (1, 1, 1) using K-means algorithm.
