Chapter 4 - Data and Analysis

Straight MCQ’s
4.1 Data Science
Part A: Knowledge-Based MCQs (Factual Recall)
Q1. Which field joins together mathematics and statistics skills with the programming
domain of computer science to extract meaningful insights from data?
A) Network Security
B) Data Science
C) Web Development
D) Computer Graphics
Correct Answer:

Q2. Who proposed the famous test in 1950 to measure a machine's ability to exhibit
intelligent behavior?
A) Alan Turing
B) Charles Babbage
C) John von Neumann
D) Ada Lovelace
Correct Answer:

Q3. Which specialized branch of Artificial Intelligence teaches computers to analyze and
derive meaningful results from digital images and videos?
A) Natural Language Processing
B) Computer Vision
C) Reinforcement Learning
D) Sentiment Analysis
Correct Answer:

Q4. What is a recommender system defined as in digital platforms?

A) A tool that erases old customer records automatically
B) A technology used to recommend products, services, or content to users based on past
behavior
C) An operating system driver for display monitors
D) A hardware firewall preventing network intrusions
Correct Answer:
Q5. What are the three primary learning models in Machine Learning?
A) Linear, Quadratic, and Exponential
B) Supervised, Unsupervised, and Reinforcement Learning
C) Tabular, Graphical, and Spatial
D) Static, Dynamic, and Hybrid
Correct Answer:

Q6. Which open-source library or framework is widely used to develop deep learning and
neural network models?
A) TensorFlow
B) MS-Word
C) SQL Server
D) HTML5
Correct Answer:

Q7. In a Supervised Machine Learning model, what kind of dataset is provided during
machine training?
A) Unlabeled raw text
B) Labeled data (questions paired with correct answers/outputs)
C) Random corrupt files
D) Non-numerical audio noise
Correct Answer:

Q8. What artificial structures are designed to function like the human brain by utilizing
interconnected artificial neurons to make decisions?
A) Logical Gates
B) Neural Networks
C) Relational Tables
D) Bus Topologies
Correct Answer:

Q9. What term describes the process of predicting whether a customer is likely to stop
using a company's services or products?
A) Churn Prediction
B) Route Optimization
C) Image Segmentation
D) Data Compression
Correct Answer:

Q10. Categorizing customers into specific groups based on their purchasing frequency,
brand loyalty, and usage habits is known as:
A) Behavioral Segmentation
B) Unsupervised Normalization
C) Data Cleaning
D) Sentiment Extraction
Correct Answer:

Q11. In Reinforcement Learning terminology, what represents the machine/computer taking

actions within an environment?
A) Agent
B) Reward
C) Penalty
D) State
Correct Answer:

Q12. What primary programming languages are most widely used in Data Science and
Machine Learning development?
A) Python and R
B) COBOL and FORTRAN
C) HTML and CSS
D) Pascal and Assembly
Correct Answer:

Part B: Understanding & Application-Based MCQs (Calculations & Logic)

Q13. How does Unsupervised Machine Learning process an input dataset during training?
A) It uses labeled answers provided by human tutors.
B) It analyzes unlabeled data to discover hidden patterns and group items based on inherent
similarities and differences.
C) It penalizes the computer every time an image is loaded.
D) It converts text into speech automatically without analysis.
Correct Answer:
Q14. If a machine learning model is trained on news articles and automatically categorizes
a new article titled "Scientists develop new vaccine" under the "Medical" category without
prior human labeling, which model was applied?
A) Supervised Learning
B) Unsupervised Learning
C) Reinforcement Learning
D) Manual Sorting
Correct Answer:

Q15. How does a Reinforcement Learning agent learn to optimize its actions over time (e.g.,
when learning to play chess)?
A) By receiving positive feedback (rewards) for helpful moves and negative feedback (penalties)
for unhelpful moves.
B) By reading a pre-labeled dataset of all possible chess games.
C) By copying the user's internet search history.
D) By translating text descriptions into audio clips.
Correct Answer:

Q16. How does automated fraud detection in banking utilize Machine Learning algorithms?
A) By identifying anomalous transaction patterns, such as sudden unusually large transfers or
irregular geographical location activity.
B) By changing bank account numbers daily.
C) By deleting user transaction histories every month.
D) By turning off ATM machines during night hours.
Correct Answer:

Q17. What distinguishes the output of a Data Science pipeline from that of an Artificial
Intelligence system?
A) Data Science produces insights, reports, and visual displays, whereas AI produces autonomous
systems that mimic intelligent human behavior.
B) Data Science generates hardware devices, while AI creates paper files.
C) Data Science works only without data, while AI requires SQL tables.
D) There is no difference between their outputs.
Correct Answer:

Q18. How do smart home voice assistants like Siri, Alexa, or Google Assistant interpret
human voice commands to turn on appliances?
A) By applying Natural Language Processing (NLP) to parse, understand, and execute spoken
speech commands.
B) By capturing thermal camera images of the room.
C) By running manual database backup scripts.
D) By measuring the air pressure inside the house.
Correct Answer:

Q19. Smartphone features like Face Unlock and Google Lens belong to which field of
Artificial Intelligence?
A) Computer Vision
B) Reinforcement Learning
C) Route Optimization
D) Behavioral Segmentation
Correct Answer:

Q20. In a machine learning workflow, why is preprocessed data split into separate "Training
Data" and "Testing Data" sets?
A) The model learns patterns from the training set, and its predictive accuracy is then evaluated
using unseen testing data.
B) Splitting data reduces the physical temperature of the CPU.
C) Testing data is used to erase errors in the training set permanently.
D) To convert numerical values into text labels.
Correct Answer:

Q21. Why is maintaining existing customer loyalty via Churn Prediction often more
beneficial for businesses than acquiring new customers?
A) Retaining existing customers costs less and yields higher long-term value than marketing to
acquire new ones.
B) New customers cannot purchase online products.
C) Existing customers do not require product shipments.
D) Churn prediction eliminates business taxes.
Correct Answer:

Q22. An airline company uses data science algorithms to adjust flight paths based on
weather patterns and passenger loads. This application is an example of:
A) Efficient route planning to reduce fuel consumption and costs
B) Computer Vision facial detection
C) Natural language sentiment analysis
D) Supervised fruit classification
Correct Answer:

Q23. How do streaming video games or music apps provide personalized

recommendations for user playlists?
A) By deploying recommender systems that evaluate the user's past listening/viewing history and
preference patterns.
B) By choosing random songs from a global database.
C) By analyzing the user's micro-processor clock speed.
D) By requiring users to vote on songs manually every hour.
Correct Answer:

Q24. What common foundation unites Data Science, Artificial Intelligence, and Machine
Learning despite their operational differences?
A) All three fields rely on a data-driven approach, utilizing algorithms and models to discover
patterns and automate decisions.
B) All three fields rely exclusively on manual paper processing.
C) All three fields operate without computer software.
D) All three fields are restricted to mechanical hardware manufacturing.
Correct Answer:

Q25. A retail store identifies that a customer hasn't visited in three months and sends a
personalized discount coupon to re-engage them. This strategy relies on:
A) Churn prediction and behavioral segmentation analysis
B) Unsupervised image rendering
C) Hardware circuit repair
D) Deep learning facial recognition
Correct Answer:

ANSWER KEY & COGNITIVE BLUEPRINT MATRIX

Item Correct
Domain Concept & Technical Principles
# Option

Data Science Definition: Intersection of math,

Q1 B Knowledge
statistics, and programming.

Turing Test: Alan Turing proposed the machine

Q2 A Knowledge
intelligence test in 1950.
Item Correct
Domain Concept & Technical Principles
# Option

Computer Vision: Visual processing branch of AI

Q3 B Knowledge
for images and videos.

Recommender Systems: User preference

Q4 B Knowledge
recommendation technology.

ML Models: Supervised, Unsupervised, and

Q5 B Knowledge
Reinforcement Learning.

Deep Learning Framework: TensorFlow supports

Q6 A Knowledge
neural network design.

Supervised Setup: Trained using pre-labeled

Q7 B Knowledge
inputs and outputs.

Neural Networks: Brain-inspired artificial neuron

Q8 B Knowledge
networks.

Churn Prediction: Identifying customers likely to

Q9 A Knowledge
cancel services.

Behavioral Segmentation: Customer grouping

Q10 A Knowledge
based on purchase habits.

Reinforcement Learning: Agent interacts with

Q11 A Knowledge
environment.

Core Languages: Python and R lead Data

Q12 A Knowledge
Science/AI development.

Unsupervised Clustering: Unlabeled pattern

Q13 B Understanding
discovery based on similarity.

Clustering Example: Text categorization via

Q14 B Application
keyphrase similarities.

Reward Mechanism: Learning via environmental

Q15 A Understanding
rewards and penalties.

Anomaly Detection: Flagging unusual financial

Q16 A Application
transactions.

Q17 A Understanding Output Comparison: Data insights vs.

 Item Correct
Domain Concept & Technical Principles
# Option

autonomous behavioral imitation.

NLP Voice Interfaces: Natural speech command

Q18 A Application
processing.

Vision Applications: Face recognition and optical

Q19 A Application
object analysis.

Model Validation: Training parameters vs. testing

Q20 A Understanding
performance evaluation.

Retention Economics: Customer retention yields

Q21 A Understanding
superior cost-efficiency.

Logistics Optimization: Route planning for fuel

Q22 A Application
and time efficiency.

Content Recommendation: Tailoring feeds

Q23 A Application
based on behavioral metrics.

Common Core: Data-driven automation and

Q24 A Understanding
predictive decision-making.

Targeted Re-engagement: Mitigating customer

Q25 A Application
churn using targeted incentives.

4.2 Data Visualization

Part A: Knowledge-Based MCQs (Factual Recall)
Q1. What term describes the process of creating graphical representations of data using
visual elements like charts, graphs, and maps?
A) Data Normalization
B) Data Visualization
C) Data Encryption
D) Data Compression
Correct Answer:

Q2. Which chart type is specifically used to display trends in data over short or long
periods of time (such as annual temperature or stock price changes)?
A) Pie Chart
B) Line Chart
C) Box Plot
D) Heatmap
Correct Answer:

Q3. Which graphical visualization represents proportions or parts of a whole across

categorical data (such as market shares)?
A) Pie Chart
B) Histogram
C) Scatter Plot
D) Bubble Chart
Correct Answer:

Q4. What visualization method displays data values using color intensity within a grid or
matrix structure?
A) Bar Chart
B) Heatmap
C) Line Chart
D) Box Plot
Correct Answer:

Q5. A Box Plot (Box-and-Whisker Plot) summarizes data distribution using five key
statistical metrics, including the minimum, maximum, median, and:
A) Mode and Range
B) First quartile and Third quartile
C) Standard deviation and Variance
D) Mean and Absolute deviation
Correct Answer:

Q6. What chart type is similar to a scatter plot but represents a third numeric variable by
modulating the size of each circular plot point?
A) Line Chart
B) Bubble Chart
C) Bar Chart
D) Histogram
Correct Answer:
Q7. Which type of data visualization specifically displays data related to physical locations,
spaces, or geographic regions?
A) Spatial Visualization
B) Temporal Visualization
C) Categorical Visualization
D) Information Visualization
Correct Answer:

Q8. What language is standardly used to execute queries for fast, reliable data retrieval in
relational database tables?
A) HTML
B) SQL
C) CSS
D) Python
Correct Answer:

Q9. Which visualization type uses continuous adjacent bars to display the frequency
distribution of a single continuous numerical variable?
A) Histogram
B) Pie Chart
C) Network Diagram
D) Tree Map
Correct Answer:

Q10. Network diagrams, tree maps, and word clouds are primary examples of which
category of visualization?
A) Quantitative Visualization
B) Information Visualization
C) Spatial Visualization
D) Temporal Visualization
Correct Answer:

Q11. Which database software applications are widely recognized relational database
management systems (RDBMS)?
A) Oracle, MS-Access, and MySQL
B) Photoshop, Illustrator, and InDesign
C) Python, R, and MATLAB
D) Word, PowerPoint, and Excel
Correct Answer:

Q12. What interface tool allows users to dynamically filter, manipulate, and explore
interactive data visualizations on digital platforms?
A) Dashboard
B) Command Line
C) Hard Drive Partition
D) Serial Port
Correct Answer:

Part B: Understanding & Application-Based MCQs (Calculations & Logic)

Q13. How does a Scatter Plot assist data analysts in examining the connection between
two continuous variables (such as student height vs. weight)?
A) By showing proportions as percentage slices of a circle.
B) By displaying individual data points across two Cartesian axes to reveal correlation patterns.
C) By converting text reviews into audio streams.
D) By arranging records in alphabetical order.
Correct Answer:

Q14. An economics research team tracks annual gold price movements over a ten-year
period. Which data visualization type is most appropriate for this task?
A) Temporal Visualization (Line Chart)
B) Spatial Visualization (Heatmap)
C) Categorical Visualization (Pie Chart)
D) Network Diagram
Correct Answer:

Q15. Why is data preprocessing (cleaning and normalization) essential before feeding
database records into Machine Learning models?
A) Machine learning algorithms require error-free, non-redundant data to prevent model
performance degradation and inaccurate predictions.
B) Preprocessing increases the physical storage size of hard drives.
C) Uncleaned data speeds up machine learning training times.
D) Preprocessing converts digital datasets into paper files.
Correct Answer:
Q16. How does Interactive Visualization enhance decision-making for business executives
compared to static print charts?
A) It enables live filtering, drill-down capabilities, and real-time trend manipulation via digital
dashboards.
B) It automatically deletes older company records.
C) It restricts data access to a single user per day.
D) It converts quantitative numbers into abstract artwork.
Correct Answer:

Q17. A public health organization plots population density variations across different
geographic provinces using color gradients on a map. What type of visualization is this?
A) Spatial Visualization
B) Temporal Line Graph
C) Multimodal Pie Chart
D) Discrete Bar Chart
Correct Answer:

Q18. How do relational databases and Machine Learning algorithms complement each
other in modern enterprise systems?
A) Databases provide structured, cleaned, and integrated data storage, which ML algorithms
retrieve and utilize for model training.
B) Databases manufacture processor chips for ML servers.
C) ML models erase database tables to free memory space.
D) Databases replace the need for machine learning algorithms.
Correct Answer:

Q19. What makes a Heatmap particularly effective for displaying complex biological
measurements, such as gene expression levels or flower part correlations?
A) Variations in color intensity instantly highlight high, low, or negative correlation values across
matrix coordinates.
B) Heatmaps require zero mathematical calculations.
C) Heatmaps convert all numerical values into text sentences.
D) Heatmaps display only three data points at a time.
Correct Answer:

Q20. If an HR department wants to compare annual salary distributions and identify salary
spread and medians across six different job roles, which plot is most effective?
A) Box Plot (Box-and-Whisker Plot)
B) Pie Chart
C) Word Cloud
D) Network Diagram
Correct Answer:

Q21. Why is a simple Bar Chart preferred over a Pie Chart when comparing discrete test
scores across five distinct students?
A) Bar charts allow clear, direct height comparisons between discrete individual categories without
relying on angle estimation.
B) Bar charts can only display two students at a time.
C) Pie charts cannot display positive numbers.
D) Bar charts turn numeric values into continuous lines.
Correct Answer:

Q22. What primary benefit does Data Visualization provide to non-technical business
stakeholders?
A) It translates complex, raw numerical datasets into clear visual stories that simplify strategic
decision-making.
B) It teaches stakeholders how to write low-level machine assembly code.
C) It replaces all human management with automated hardware switches.
D) It removes all graphical elements from corporate presentations.
Correct Answer:

Q23. Studying relationships between three variables simultaneously—such as marketing

ad duration (X), campaign effectiveness (Y), and total budget spent (Z)—is best achieved
using a:
A) Multivariate Visualization (Bubble Chart / Scatter Matrix)
B) Simple 2D Line Graph
C) Standard Pie Chart
D) Single Column Table
Correct Answer:

Q24. In terms of scalability, how do modern Database Management Systems (DBMS)

support large-scale Machine Learning applications?
A) DBMS platforms can scale from handling small local tables up to managing millions of national
records required for Big Data training sets.
B) DBMS platforms restrict datasets to maximum 100 records.
C) DBMS platforms convert data into analog television signals.
D) DBMS platforms eliminate the need for server memory.
Correct Answer:

Q25. How does a sports analytics team use Statistical Visualization (like histograms and
scatter plots) to improve player performance?
A) By examining player metric distributions and correlations between physical training loads and
match performance.
B) By changing the rules of the sport automatically.
C) By printing static poster images of team logos.
D) By converting game video files into text transcripts only.
Correct Answer:

ANSWER KEY & COGNITIVE BLUEPRINT MATRIX

Item Correct
Domain Concept & Technical Principles
# Option

Data Visualization Definition: Graphical

Q1 B Knowledge
representation of data via charts and maps.

Line Chart Purpose: Displays numerical data

Q2 B Knowledge
trends over time.

Pie Chart Purpose: Shows proportional

Q3 A Knowledge
breakdown of a whole category set.

Heatmap Definition: Matrix data representation

Q4 B Knowledge
using color intensity scales.

Box Plot Metrics: Minimum, maximum, median,

Q5 B Knowledge
Q1, and Q3 boundaries.

Bubble Chart: Extends scatter plots by introducing

Q6 B Knowledge
a 3rd variable as bubble area.

Spatial Visualization: Visualizes geographic or

Q7 A Knowledge
location-bound physical data.

SQL Standard: Structured Query Language

Q8 B Knowledge
manages relational database queries.

Histogram Definition: Displays frequency

Q9 A Knowledge
distribution of a continuous variable.
Item Correct
Domain Concept & Technical Principles
# Option

Information Visualization: Presents complex

Q10 B Knowledge
abstract/conceptual structures.

RDBMS Examples: Oracle, MS-Access, and

Q11 A Knowledge
MySQL manage structured tables.

Dashboard Tool: Digital interface facilitating

Q12 A Knowledge
dynamic chart interaction.

Scatter Plot Analysis: Evaluates correlation

Q13 B Understanding
between two continuous variables.

Temporal Analytics: Tracking time-series trend

Q14 A Application
lines for longitudinal data.

Data Quality in ML: Clean, normalized input

Q15 A Understanding
prevents algorithmic bias and error.

Interactive Dashboards: Dynamic filtering allows

Q16 A Application
deeper exploration of insights.

Geographic Spatial Mapping: Representing

Q17 A Application
regional data via thematic maps.

Database-ML Synergy: Databases supply

Q18 A Understanding
structured data pipelines for ML models.

Heatmap Color Gradients: Expresses numerical

Q19 A Understanding
intensity/correlation across dimensions.

Distribution Comparison: Box plots show spread,

Q20 A Application
median, and quartile variation.

Discrete Category Comparison: Bar heights offer

Q21 A Understanding
direct visual precision over pie angles.

Stakeholder Value: Translates complex

Q22 A Understanding
mathematical findings into accessible insights.

Multivariate Analysis: Bubble charts project three

Q23 A Application
variable dimensions onto 2D space.

Q24 A Understanding DBMS Scalability: Efficiently scales data storage

 Item Correct
Domain Concept & Technical Principles
# Option

capacity to feed ML models.

Sports Analytics: Identifying statistical trends and

Q25 A Application
distribution performance bounds

4.3 Stages of the Data Science Life Cycle

Part A: Knowledge-Based MCQs (Factual Recall)
Q1. Which statement best characterizes the overall nature of the Data Science Life Cycle?
A) It is a rigid, one-time linear path with no repetition.
B) It is an iterative, structured process with interconnected stages to solve data-driven problems.
C) It operates exclusively without human input or supervision.
D) It is a hardware manufacturing workflow.
Correct Answer:

Q2. What are Key Performance Indicators (KPIs) in the Problem Definition stage?
A) Hardware components used to store databases
B) Measurable targets defined to track the progress and success of proposed solutions
C) Uncleaned raw data files
D) Encryption keys used for web security
Correct Answer:

Q3. What software mechanism acts like a digital messenger between different applications
to retrieve and exchange data automatically?
A) Central Processing Unit (CPU)
B) Application Programming Interface (API)
C) Graphical User Interface (GUI)
D) Read-Only Memory (ROM)
Correct Answer:

Q4. Data cleaning is also commonly known as:

A) Data preprocessing
B) Model deployment
C) Route optimization
D) Predictive rendering
Correct Answer:

Q5. During which stage of the data science life cycle are errors, duplicate records, and
missing values resolved?
A) Data Collection
B) Data Cleaning
C) Model Evaluation
D) Maintenance and Monitoring
Correct Answer:

Q6. What primary outcome is produced at the completion of the Problem Definition stage?
A) A deployed web application
B) A hypothesis or well-defined problem statement
C) A trained deep learning model
D) An unorganized SQL table
Correct Answer:

Q7. What stage immediately follows Data Cleaning in the standard Data Science Life Cycle?
A) Maintenance and Monitoring
B) Data Analysis
C) Model Deployment
D) API Integration
Correct Answer:

Q8. Which stage focuses on setting up databases and building structured diagrams that
show relationships between entities and attributes?
A) Data Modeling
B) Problem Definition
C) Model Evaluation
D) Data Collection
Correct Answer:

Q9. Measuring a model's performance against pre-established metrics like accuracy, speed,
and cost-effectiveness occurs in which stage?
A) Model Evaluation
B) Data Collection
C) Data Cleaning
D) Problem Definition
Correct Answer:

Q10. Integrating a trained machine learning model directly into an active website, mobile
application, or inventory database is known as:
A) Model Evaluation
B) Model Deployment
C) Data Extraction
D) Problem Scoping
Correct Answer:

Q11. What ongoing stage ensures that a machine learning system maintains accuracy and
speed after it has been launched into production?
A) Data Modeling
B) Maintenance and Monitoring
C) Problem Definition
D) Data Collection
Correct Answer:

Q12. In a retail data science project, tracking current stock levels, historical sales rates,
and supplier price promotional changes is part of:
A) Data Collection
B) Model Deployment
C) Algorithm Selection
D) System Formatting
Correct Answer:

Part B: Understanding & Application-Based MCQs (Calculations & Logic)

Q13. A company sets a goal to "increase online sales revenue by 15% within two months."
This statement represents a:
A) Key Performance Indicator (KPI) within the Problem Definition stage
B) Data cleaning error
C) Machine learning algorithm parameter
D) Hardware storage specification
Correct Answer:
Q14. If a dataset contains customer birth dates, creating a new attribute column for "Day of
the Week" during data preprocessing is an example of:
A) Feature engineering during Data Cleaning
B) Model deployment failure
C) Problem definition scoping
D) Database corruption
Correct Answer:

Q15. Why is the quality and accuracy of collected raw data critical to the overall success of
a data science project?
A) Inaccurate input data directly degrades the reliability of insights and predictions in later stages.
B) Low-quality data destroys system hard drives.
C) High-quality data eliminates the need for software programming.
D) Raw data quality affects only the internet connection speed.
Correct Answer:

Q16. What step should a data science team take if a machine learning model fails to meet
target accuracy thresholds during the Model Evaluation stage?
A) Immediately deploy the failed model to production.
B) Return to the Data Modeling stage to adjust algorithms and fine-tune parameters.
C) Delete all original historical data permanently.
D) Stop the project completely without revision.
Correct Answer:

Q17. A retail superstore uses data analysis to answer: "Which smartphone models sell
fastest, and which remain unsold on shelves for too long?" This activity directly serves to:
A) Identify stock demand trends to optimize future reordering schedules.
B) Increase the physical weight of retail products.
C) Replace digital payment gateways with cash-only systems.
D) Permanently close slow-performing store branches.
Correct Answer:

Q18. How does partial (phased) deployment of a model benefit an enterprise software
system compared to immediate full deployment?
A) It allows testing system performance and stability on a small subset of real-world data before
scaling across the entire organization.
B) It guarantees that the system will never require maintenance updates.
C) It eliminates the need for user authentication.
D) It reduces server electricity consumption to zero.
Correct Answer:

Q19. During smartphone inventory modeling, what triggers the automated inventory
system to signal a reorder request?
A) Rapid drops in stock levels detected through real-time sales tracking.
B) A manual server restart by system administrators.
C) Changes in room lighting inside the warehouse.
D) Deletion of supplier contact records.
Correct Answer:

Q20. Why must data privacy and security compliance be strictly evaluated alongside model
accuracy in the Model Evaluation stage?
A) To prevent unauthorized exposure or misuse of sensitive user and customer information.
B) To speed up the compilation time of Python code.
C) To convert structured SQL tables into unstructured text files.
D) To allow public access to all confidential database records.
Correct Answer:

Q21. If a store's sales management software begins making inaccurate inventory

predictions due to sudden shifts in customer buying habits, which life cycle stage
addresses this through retraining?
A) Maintenance and Monitoring
B) Data Collection
C) Initial Scoping
D) API Design
Correct Answer:

Q22. An investigator cross-references missing product price entries in a store's database

with supplier records to fill in missing values. This action falls under:
A) Handling missing values in Data Cleaning
B) Model Deployment
C) Problem Definition
D) System Architecture Design
Correct Answer:
Q23. How do data visualization tools (such as line charts and bar graphs) assist analysts
during the Data Analysis stage?
A) They present complex data trends visually, making seasonal purchasing spikes easier to detect.
B) They automatically write executable source code for machine learning algorithms.
C) They compress database files into zip archives.
D) They encrypt sensitive passwords stored in database fields.
Correct Answer:

Q24. A superstore balances inventory to prevent running out of popular smartphone

models while simultaneously avoiding overstocking slow-moving items. This dual goal
optimizes:
A) Cost-effectiveness and inventory efficiency
B) Hardware clock speeds
C) Manual data entry speeds
D) Unstructured network bandwidth
Correct Answer:

Q25. Which sequence correctly reflects the chronological flow of the Data Science Life
Cycle stages?
A) Problem Definition → Data Collection → Data Cleaning → Data Analysis → Data Modeling →
Model Evaluation → Model Deployment → Maintenance & Monitoring
B) Model Deployment → Data Analysis → Problem Definition → Data Collection → Model
Evaluation
C) Maintenance & Monitoring → Data Cleaning → Data Modeling → Model Deployment →
Problem Definition
D) Data Modeling → Problem Definition → Data Collection → Model Deployment → Data
Cleaning
Correct Answer:

ANSWER KEY & COGNITIVE BLUEPRINT MATRIX

Item Correct
Domain Concept & Technical Principles
# Option

Life Cycle Characterization: Iterative, structured

Q1 B Knowledge
problem-solving approach.

KPI Definition: Quantifiable metrics designed to

Q2 B Knowledge
evaluate progress and success.
Item Correct
Domain Concept & Technical Principles
# Option

API Definition: Interface mechanism facilitating

Q3 B Knowledge
inter-application data exchange.

Terminology: Data cleaning is synonymous with

Q4 A Knowledge
data preprocessing.

Data Cleaning Purpose: Error correction,

Q5 B Knowledge
deduplication, and missing value handling.

Problem Definition Output: Formulation of clear

Q6 B Knowledge
objectives and well-defined hypotheses.

Life Cycle Flow: Data Analysis succeeds Data

Q7 B Knowledge
Cleaning.

Data Modeling: Establishing structural schemas,

Q8 A Knowledge
entity relationships, and databases.

Model Evaluation: Benchmarking performance

Q9 A Knowledge
against accuracy and efficiency metrics.

Deployment Definition: Integration of trained

Q10 B Knowledge
models into production environments.

Maintenance Role: Post-deployment tracking of

Q11 B Knowledge
operational accuracy and speed.

Case Study Collection: Gathering inventory

Q12 A Knowledge
parameters and sales tracking metrics.

KPI Target Setting: Quantifiable sales growth

Q13 A Application
target over a specified period.

Feature Engineering: Creating derived attributes

Q14 A Understanding
during preprocessing.

Data Quality Impact: Garbage-in/garbage-out

Q15 A Understanding
principle in data analytics.

Iterative Loop: Re-modeling and tuning upon

Q16 B Understanding
evaluation metric shortfall.

Q17 A Application Analytical Insight: Driving inventory reordering

Item Correct
Domain Concept & Technical Principles
# Option

decisions via sales trends.

Phased Deployment: Mitigating operational risk

Q18 A Understanding
through incremental rollout.

Automated Reordering: Inventory logic

Q19 A Understanding
responding to demand velocity.

Governance & Ethics: Safeguarding sensitive

Q20 A Understanding
personal and organizational data.

Model Drift Management: Retraining models as

Q21 A Application
underlying data distributions shift.

Imputation Strategy: Resolving missing attributes

Q22 A Application
via source cross-referencing.

Exploratory Visualization: Enhancing trend

Q23 A Understanding
identification via graphical plots.

Business Optimization: Balancing stock

Q24 A Application
availability against carrying costs.

Sequential Workflow: End-to-end lifecycle

Q25 A Knowledge
progression mapping.