ISYE 6501 Final Exam Review | Analytics Modeling Practice
Examination
Advanced Comprehensive Practice Test
Exam Title: ISYE 6501 Final Examination: Advanced Analytics Modeling Practice Assessment Covering
Supervised and Unsupervised Learning, Time Series Analysis, Optimization, Simulation, and Model
Validation Techniques for Graduate-Level Analytics
SECTION 1: SUPERVISED VS. UNSUPERVISED LEARNING (Questions 1-
15)
Question 1:
Which of the following is an example of a supervised learning problem?
A) Grouping customers into segments based on purchasing behavior
B) Predicting house prices based on square footage and number of bedrooms
C) Reducing the dimensionality of a dataset while preserving variance
D) Finding association rules in market basket data
Correct Answer: B
Rationale: Supervised learning uses labeled data to predict an outcome. Predicting house prices (a
continuous outcome) from features is supervised regression. Clustering (A) and association rules (D)
are unsupervised methods, while PCA (C) is dimensionality reduction .
Question 2:
Which of the following is an example of an unsupervised learning task?
A) Predicting whether an email is spam or not
B) Estimating house prices
C) Grouping news articles into topics
D) Forecasting stock prices
Correct Answer: C
Rationale: Topic modeling (grouping articles into topics without predefined labels) is unsupervised
learning. Spam detection, price prediction, and stock forecasting are supervised tasks requiring labeled
historical data .
,Question 3:
What do descriptive questions in analytics typically ask?
A) What action(s) would be best?
B) What will happen?
C) What happened?
D) How can we optimize outcomes?
Correct Answer: C
Rationale: Descriptive questions focus on understanding past events and patterns, such as "which
customers are most alike?" This contrasts with predictive questions (what will happen?) and
prescriptive questions (what actions should be taken?) .
Question 4:
What do predictive questions in analytics typically ask?
A) What happened?
B) What will happen?
C) What action(s) would be best?
D) Which customers are most alike?
Correct Answer: B
Rationale: Predictive questions focus on forecasting future outcomes, such as "what will Google's
stock price be?" This is distinct from descriptive questions (what happened?) and prescriptive
questions (what actions should be taken?) .
Question 5:
What do prescriptive questions in analytics typically ask?
A) What happened?
B) What will happen?
C) What action(s) would be best?
D) Which customers are most similar?
Correct Answer: C
Rationale: Prescriptive questions focus on determining optimal actions, such as "where to put traffic
lights?" This represents the most advanced stage of analytics, recommending decisions based on
predictive insights .
,Question 6:
A company wants to predict which customers are likely to churn based on historical data about
customers who previously churned or stayed. This is an example of:
A) Unsupervised learning (clustering)
B) Supervised learning (classification)
C) Dimensionality reduction
D) Association rule mining
Correct Answer: B
Rationale: The company has labeled historical data (churned vs. stayed) and wants to predict future
churn for new customers, making this a supervised binary classification problem .
Question 7:
An unsupervised learning algorithm that reduces data dimensionality is:
A) Linear regression
B) Logistic regression
C) Principal Component Analysis (PCA)
D) Decision tree
Correct Answer: C
Rationale: PCA is an unsupervised dimensionality reduction technique that projects data onto
orthogonal components ordered by explained variance. It does not use labels, distinguishing it from
regression and decision tree methods .
Question 8:
In supervised learning, what distinguishes classification from regression?
A) Classification uses continuous inputs; regression uses categorical inputs
B) Classification predicts discrete/categorical outcomes; regression predicts continuous outcomes
C) Classification is unsupervised; regression is supervised
D) Classification only works with two classes
Correct Answer: B
Rationale: Classification predicts a discrete class label (e.g., spam/not spam, digit 0-9), while regression
predicts a continuous numeric value (e.g., house price, temperature) .
, Question 9:
Which of the following is a classification problem?
A) Predicting tomorrow's temperature in degrees Fahrenheit
B) Estimating the selling price of a used car
C) Determining whether an email is spam or legitimate
D) Forecasting next month's sales revenue
Correct Answer: C
Rationale: Spam detection involves predicting a binary category (spam or legitimate), making it a
classification problem. The others involve predicting continuous values, making them regression
problems .
Question 10:
Which of the following is a regression problem?
A) Identifying handwritten digits (0-9)
B) Classifying iris flower species
C) Predicting a student's GPA based on study hours
D) Determining if a transaction is fraudulent
Correct Answer: C
Rationale: GPA is a continuous numeric value (e.g., 3.75, 2.89), making this a regression problem. The
others involve predicting discrete categories (classification) .
Question 11:
Semi-supervised learning is most valuable when:
A) Labeled data is abundant and cheap
B) Unlabeled data is abundant but labeled data is expensive or time-consuming to obtain
C) Both labeled and unlabeled data are equally available
D) Neither labeled nor unlabeled data is available
Correct Answer: B
Rationale: Semi-supervised learning is useful when labeling is costly (e.g., medical images requiring
expert annotation) but large amounts of unlabeled data are available. It combines a small amount of
labeled data with a large amount of unlabeled data during training .
Examination
Advanced Comprehensive Practice Test
Exam Title: ISYE 6501 Final Examination: Advanced Analytics Modeling Practice Assessment Covering
Supervised and Unsupervised Learning, Time Series Analysis, Optimization, Simulation, and Model
Validation Techniques for Graduate-Level Analytics
SECTION 1: SUPERVISED VS. UNSUPERVISED LEARNING (Questions 1-
15)
Question 1:
Which of the following is an example of a supervised learning problem?
A) Grouping customers into segments based on purchasing behavior
B) Predicting house prices based on square footage and number of bedrooms
C) Reducing the dimensionality of a dataset while preserving variance
D) Finding association rules in market basket data
Correct Answer: B
Rationale: Supervised learning uses labeled data to predict an outcome. Predicting house prices (a
continuous outcome) from features is supervised regression. Clustering (A) and association rules (D)
are unsupervised methods, while PCA (C) is dimensionality reduction .
Question 2:
Which of the following is an example of an unsupervised learning task?
A) Predicting whether an email is spam or not
B) Estimating house prices
C) Grouping news articles into topics
D) Forecasting stock prices
Correct Answer: C
Rationale: Topic modeling (grouping articles into topics without predefined labels) is unsupervised
learning. Spam detection, price prediction, and stock forecasting are supervised tasks requiring labeled
historical data .
,Question 3:
What do descriptive questions in analytics typically ask?
A) What action(s) would be best?
B) What will happen?
C) What happened?
D) How can we optimize outcomes?
Correct Answer: C
Rationale: Descriptive questions focus on understanding past events and patterns, such as "which
customers are most alike?" This contrasts with predictive questions (what will happen?) and
prescriptive questions (what actions should be taken?) .
Question 4:
What do predictive questions in analytics typically ask?
A) What happened?
B) What will happen?
C) What action(s) would be best?
D) Which customers are most alike?
Correct Answer: B
Rationale: Predictive questions focus on forecasting future outcomes, such as "what will Google's
stock price be?" This is distinct from descriptive questions (what happened?) and prescriptive
questions (what actions should be taken?) .
Question 5:
What do prescriptive questions in analytics typically ask?
A) What happened?
B) What will happen?
C) What action(s) would be best?
D) Which customers are most similar?
Correct Answer: C
Rationale: Prescriptive questions focus on determining optimal actions, such as "where to put traffic
lights?" This represents the most advanced stage of analytics, recommending decisions based on
predictive insights .
,Question 6:
A company wants to predict which customers are likely to churn based on historical data about
customers who previously churned or stayed. This is an example of:
A) Unsupervised learning (clustering)
B) Supervised learning (classification)
C) Dimensionality reduction
D) Association rule mining
Correct Answer: B
Rationale: The company has labeled historical data (churned vs. stayed) and wants to predict future
churn for new customers, making this a supervised binary classification problem .
Question 7:
An unsupervised learning algorithm that reduces data dimensionality is:
A) Linear regression
B) Logistic regression
C) Principal Component Analysis (PCA)
D) Decision tree
Correct Answer: C
Rationale: PCA is an unsupervised dimensionality reduction technique that projects data onto
orthogonal components ordered by explained variance. It does not use labels, distinguishing it from
regression and decision tree methods .
Question 8:
In supervised learning, what distinguishes classification from regression?
A) Classification uses continuous inputs; regression uses categorical inputs
B) Classification predicts discrete/categorical outcomes; regression predicts continuous outcomes
C) Classification is unsupervised; regression is supervised
D) Classification only works with two classes
Correct Answer: B
Rationale: Classification predicts a discrete class label (e.g., spam/not spam, digit 0-9), while regression
predicts a continuous numeric value (e.g., house price, temperature) .
, Question 9:
Which of the following is a classification problem?
A) Predicting tomorrow's temperature in degrees Fahrenheit
B) Estimating the selling price of a used car
C) Determining whether an email is spam or legitimate
D) Forecasting next month's sales revenue
Correct Answer: C
Rationale: Spam detection involves predicting a binary category (spam or legitimate), making it a
classification problem. The others involve predicting continuous values, making them regression
problems .
Question 10:
Which of the following is a regression problem?
A) Identifying handwritten digits (0-9)
B) Classifying iris flower species
C) Predicting a student's GPA based on study hours
D) Determining if a transaction is fraudulent
Correct Answer: C
Rationale: GPA is a continuous numeric value (e.g., 3.75, 2.89), making this a regression problem. The
others involve predicting discrete categories (classification) .
Question 11:
Semi-supervised learning is most valuable when:
A) Labeled data is abundant and cheap
B) Unlabeled data is abundant but labeled data is expensive or time-consuming to obtain
C) Both labeled and unlabeled data are equally available
D) Neither labeled nor unlabeled data is available
Correct Answer: B
Rationale: Semi-supervised learning is useful when labeling is costly (e.g., medical images requiring
expert annotation) but large amounts of unlabeled data are available. It combines a small amount of
labeled data with a large amount of unlabeled data during training .
