ISYE 6501 Final Quiz - 200 Questions &
Answers with Rationales
Winter 2026 | Georgia Tech | 100% Verified
SECTION 1: FOUNDATIONAL CONCEPTS (Questions 1-25)
Question 1
What do descriptive questions ask?
A) What will happen?
B) What action would be best?
C) What happened?
D) How to optimize a process?
Answer: C) What happened?
Rationale: Descriptive analytics focuses on summarizing historical data to
understand what has occurred. Examples include identifying which customers are
most alike or describing sales patterns. Descriptive questions establish the baseline
understanding before moving to predictive or prescriptive analysis .
Question 2
What do predictive questions ask?
A) What happened?
B) What will happen?
C) What action would be best?
D) How to describe the data?
Answer: B) What will happen?
,Rationale: Predictive analytics uses historical data to forecast future outcomes.
Examples include predicting stock prices, customer churn, or sales volumes.
Predictive models identify patterns in past data to estimate future events .
Question 3
What do prescriptive questions ask?
A) What happened?
B) What will happen?
C) What action(s) would be best?
D) How to visualize the data?
Answer: C) What action(s) would be best?
Rationale: Prescriptive analytics recommends optimal actions based on
predictions. It combines predictive insights with optimization to guide decision-
making. Examples include determining optimal traffic light placement or resource
allocation.
Question 4
What is a model in the context of analytics?
A) A physical representation of data
B) A real-life situation expressed as mathematics
C) A type of algorithm only
D) A data visualization tool
Answer: B) A real-life situation expressed as mathematics
Rationale: A model simplifies reality to capture essential relationships
mathematically. It enables predictions, decision support, and understanding of
complex systems through mathematical representation .
Question 5
,What do classifiers help you do?
A) Predict continuous values
B) Differentiate and categorize data into classes
C) Cluster similar data points
D) Reduce dimensionality
Answer: B) Differentiate and categorize data into classes
Rationale: Classifiers assign data points to discrete categories or classes based on
learned patterns. They differentiate between groups (e.g., spam vs. not spam, fraud
vs. legitimate).
Question 6
What is a soft classifier and when is it used?
A) A classifier with 100% accuracy
B) A classifier that minimizes mistakes when perfect separation doesn't exist
C) A classifier that only works with continuous data
D) A classifier that always predicts the majority class
Answer: B) A classifier that minimizes mistakes when perfect separation
doesn't exist
Rationale: In real-world data, perfect separation often doesn't exist. Soft classifiers
optimize the trade-off between accuracy and margin, accepting some errors to
achieve optimal overall performance .
Question 7
What is a data point?
A) The average value in a dataset
B) A single row in a data table
C) A column header
D) The total number of observations
Answer: B) A single row in a data table
, Rationale: A data point (or record) is an individual observation in a dataset—one
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
row representing a single entity with its associated attributes .
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
Question 8 k0
What is an attribute/feature/covariate?
k0 k0 k0
A) A row in a data table
k0 k0 k0 k0 k0
B) A column in a data table
k0 k0 k0 k0 k0
C) The response variable k0 k0
D) The model output k0 k0
Answer: B) A column in a data table k0 k0 k0 k0 k0 k0 k0
Rationale: Attributes (also called features, covariates, predictors, or factors) are
k0 k0 k0 k0 k0 k0 k0 k0 k0
the characteristics or variables measured for each data point. They are represented
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
as columns .
k0 k0 k0
Question 9 k0
What is the response/outcome variable?
k0 k0 k0 k0
A) A predictor variable
k0 k0
B) The 'answer' for each data point
k0 k0 k0 k0 k0
C) A categorical feature
k0 k0
D) An input to the model k0 k0 k0 k0
Answer: B) The 'answer' for each data point
k0 k0 k0 k0 k0 k0 k0
Rationale: The response (or outcome) variable is what the model aims to predict.
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
It is the dependent variable in regression or the class label in classification
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
problems .
k0 k0
Question 10 k0
What is time-series data? k0 k0 k0
Answers with Rationales
Winter 2026 | Georgia Tech | 100% Verified
SECTION 1: FOUNDATIONAL CONCEPTS (Questions 1-25)
Question 1
What do descriptive questions ask?
A) What will happen?
B) What action would be best?
C) What happened?
D) How to optimize a process?
Answer: C) What happened?
Rationale: Descriptive analytics focuses on summarizing historical data to
understand what has occurred. Examples include identifying which customers are
most alike or describing sales patterns. Descriptive questions establish the baseline
understanding before moving to predictive or prescriptive analysis .
Question 2
What do predictive questions ask?
A) What happened?
B) What will happen?
C) What action would be best?
D) How to describe the data?
Answer: B) What will happen?
,Rationale: Predictive analytics uses historical data to forecast future outcomes.
Examples include predicting stock prices, customer churn, or sales volumes.
Predictive models identify patterns in past data to estimate future events .
Question 3
What do prescriptive questions ask?
A) What happened?
B) What will happen?
C) What action(s) would be best?
D) How to visualize the data?
Answer: C) What action(s) would be best?
Rationale: Prescriptive analytics recommends optimal actions based on
predictions. It combines predictive insights with optimization to guide decision-
making. Examples include determining optimal traffic light placement or resource
allocation.
Question 4
What is a model in the context of analytics?
A) A physical representation of data
B) A real-life situation expressed as mathematics
C) A type of algorithm only
D) A data visualization tool
Answer: B) A real-life situation expressed as mathematics
Rationale: A model simplifies reality to capture essential relationships
mathematically. It enables predictions, decision support, and understanding of
complex systems through mathematical representation .
Question 5
,What do classifiers help you do?
A) Predict continuous values
B) Differentiate and categorize data into classes
C) Cluster similar data points
D) Reduce dimensionality
Answer: B) Differentiate and categorize data into classes
Rationale: Classifiers assign data points to discrete categories or classes based on
learned patterns. They differentiate between groups (e.g., spam vs. not spam, fraud
vs. legitimate).
Question 6
What is a soft classifier and when is it used?
A) A classifier with 100% accuracy
B) A classifier that minimizes mistakes when perfect separation doesn't exist
C) A classifier that only works with continuous data
D) A classifier that always predicts the majority class
Answer: B) A classifier that minimizes mistakes when perfect separation
doesn't exist
Rationale: In real-world data, perfect separation often doesn't exist. Soft classifiers
optimize the trade-off between accuracy and margin, accepting some errors to
achieve optimal overall performance .
Question 7
What is a data point?
A) The average value in a dataset
B) A single row in a data table
C) A column header
D) The total number of observations
Answer: B) A single row in a data table
, Rationale: A data point (or record) is an individual observation in a dataset—one
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
row representing a single entity with its associated attributes .
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
Question 8 k0
What is an attribute/feature/covariate?
k0 k0 k0
A) A row in a data table
k0 k0 k0 k0 k0
B) A column in a data table
k0 k0 k0 k0 k0
C) The response variable k0 k0
D) The model output k0 k0
Answer: B) A column in a data table k0 k0 k0 k0 k0 k0 k0
Rationale: Attributes (also called features, covariates, predictors, or factors) are
k0 k0 k0 k0 k0 k0 k0 k0 k0
the characteristics or variables measured for each data point. They are represented
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
as columns .
k0 k0 k0
Question 9 k0
What is the response/outcome variable?
k0 k0 k0 k0
A) A predictor variable
k0 k0
B) The 'answer' for each data point
k0 k0 k0 k0 k0
C) A categorical feature
k0 k0
D) An input to the model k0 k0 k0 k0
Answer: B) The 'answer' for each data point
k0 k0 k0 k0 k0 k0 k0
Rationale: The response (or outcome) variable is what the model aims to predict.
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
It is the dependent variable in regression or the class label in classification
k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0 k0
problems .
k0 k0
Question 10 k0
What is time-series data? k0 k0 k0
