1
SYE 6501 Intro to Analytics Modeling Practice Exam: 100 Questions with Answers
Section 1: Classification Models — SVM, k-NN, CART (Questions 1–25)
Q1. What does SVM stand for?
A. Statistical Vector Modeling
B. Support Vector Machine
C. Supervised Variable Method
D. Sequential Validation Model
Answer: B
Rationale: SVM stands for Support Vector Machine .
Q2. In an SVM soft classifier, what is the tradeoff?
A. Maximizing margin and minimizing errors
B. Minimizing margin and maximizing errors
C. Maximizing both margin and errors
D. Minimizing both margin and errors
Answer: A
Rationale: In a soft SVM classifier, we trade off between maximizing the margin and
minimizing classification errors .
Q3. If the lambda parameter gets small, what gets emphasized in an SVM?
A. Large margin
B. Minimizing training error
C. Maximizing margin
D. Removing outliers
Answer: B
Rationale: When lambda gets small, minimizing errors is emphasized rather than maximizing
the margin .
Q4. What is a support vector?
A. A data point that holds up the margin
B. A vector of coefficients
C. The center of a cluster
D. A type of kernel function
Answer: A
Rationale: A support vector is a data point that "holds up" the shape of the classifier .
Q5. If an SVM coefficient is very close to zero, what does this indicate?
, 2
A. The term is very important for classification
B. The term is not very important for classification
C. The term should be removed
D. The term causes overfitting
Answer: B
Rationale: A coefficient close to zero indicates that term is not very important to the
classification .
Q6. Why is scaling or normalizing data critical for k-NN?
A. To reduce computational complexity
B. To prevent attributes with larger ranges from dominating distance calculations
C. To transform data into a normal distribution
D. To handle categorical variables
Answer: B
Rationale: Without scaling, attributes with larger numerical ranges dominate the distance
calculation in k-NN .
Q7. What is the difference between standardization and scaling?
A. Scaling is bounded in range; standardization transforms to a normal distribution
B. They are the same thing
C. Standardization is bounded; scaling is not
D. Scaling removes outliers; standardization does not
Answer: A
Rationale: Scaling is bounded in range, while standardization scales to a normal distribution
using (value - mean) / standard deviation .
Q8. What does a confusion matrix show?
A. The performance of a classification model
B. The distribution of continuous variables
C. The correlation between features
D. The time series trend
Answer: A
Rationale: A confusion matrix shows the performance of a classification model .
Q9. In a regression tree (CART), how is pruning typically conducted?
A. By removing branches that decrease variance in the training set
B. By using validation data to see if branching improves estimation error
C. By increasing tree depth until each leaf has one point
D. By replacing quantitative variables with binary ones
, 3
Answer: B
Rationale: Pruning uses the second half of the data (validation) to determine if branching
actually improves estimation error .
Q10. What is the primary benefit of a Random Forest compared to a single Decision Tree?
A. Reduction in variance and improved accuracy through averaging
B. Faster computation time
C. Higher interpretability
D. Simpler implementation
Answer: A
Rationale: Random Forest reduces variance and improves accuracy by averaging multiple
decision trees .
Q11. What is the 2-norm distance?
A. Euclidean distance
B. Manhattan distance
C. The largest dimension value
D. The sum of absolute values
Answer: A
Rationale: The 2-norm is Euclidean (straight-line) distance .
Q12. What is the 1-norm distance?
A. Euclidean distance
B. Rectilinear (Manhattan) distance
C. The largest dimension value
D. The square root of the sum of squares
Answer: B
Rationale: The 1-norm is the rectilinear (Manhattan) distance .
Q13. What is the infinity norm?
A. Euclidean distance
B. Manhattan distance
C. The value of the largest dimension
D. The sum of all dimensions
Answer: C
Rationale: The infinity norm is the value of the largest dimension .
Q14. In a Support Vector Machine, if the parameter C is set very large, what is most likely to
occur?
SYE 6501 Intro to Analytics Modeling Practice Exam: 100 Questions with Answers
Section 1: Classification Models — SVM, k-NN, CART (Questions 1–25)
Q1. What does SVM stand for?
A. Statistical Vector Modeling
B. Support Vector Machine
C. Supervised Variable Method
D. Sequential Validation Model
Answer: B
Rationale: SVM stands for Support Vector Machine .
Q2. In an SVM soft classifier, what is the tradeoff?
A. Maximizing margin and minimizing errors
B. Minimizing margin and maximizing errors
C. Maximizing both margin and errors
D. Minimizing both margin and errors
Answer: A
Rationale: In a soft SVM classifier, we trade off between maximizing the margin and
minimizing classification errors .
Q3. If the lambda parameter gets small, what gets emphasized in an SVM?
A. Large margin
B. Minimizing training error
C. Maximizing margin
D. Removing outliers
Answer: B
Rationale: When lambda gets small, minimizing errors is emphasized rather than maximizing
the margin .
Q4. What is a support vector?
A. A data point that holds up the margin
B. A vector of coefficients
C. The center of a cluster
D. A type of kernel function
Answer: A
Rationale: A support vector is a data point that "holds up" the shape of the classifier .
Q5. If an SVM coefficient is very close to zero, what does this indicate?
, 2
A. The term is very important for classification
B. The term is not very important for classification
C. The term should be removed
D. The term causes overfitting
Answer: B
Rationale: A coefficient close to zero indicates that term is not very important to the
classification .
Q6. Why is scaling or normalizing data critical for k-NN?
A. To reduce computational complexity
B. To prevent attributes with larger ranges from dominating distance calculations
C. To transform data into a normal distribution
D. To handle categorical variables
Answer: B
Rationale: Without scaling, attributes with larger numerical ranges dominate the distance
calculation in k-NN .
Q7. What is the difference between standardization and scaling?
A. Scaling is bounded in range; standardization transforms to a normal distribution
B. They are the same thing
C. Standardization is bounded; scaling is not
D. Scaling removes outliers; standardization does not
Answer: A
Rationale: Scaling is bounded in range, while standardization scales to a normal distribution
using (value - mean) / standard deviation .
Q8. What does a confusion matrix show?
A. The performance of a classification model
B. The distribution of continuous variables
C. The correlation between features
D. The time series trend
Answer: A
Rationale: A confusion matrix shows the performance of a classification model .
Q9. In a regression tree (CART), how is pruning typically conducted?
A. By removing branches that decrease variance in the training set
B. By using validation data to see if branching improves estimation error
C. By increasing tree depth until each leaf has one point
D. By replacing quantitative variables with binary ones
, 3
Answer: B
Rationale: Pruning uses the second half of the data (validation) to determine if branching
actually improves estimation error .
Q10. What is the primary benefit of a Random Forest compared to a single Decision Tree?
A. Reduction in variance and improved accuracy through averaging
B. Faster computation time
C. Higher interpretability
D. Simpler implementation
Answer: A
Rationale: Random Forest reduces variance and improves accuracy by averaging multiple
decision trees .
Q11. What is the 2-norm distance?
A. Euclidean distance
B. Manhattan distance
C. The largest dimension value
D. The sum of absolute values
Answer: A
Rationale: The 2-norm is Euclidean (straight-line) distance .
Q12. What is the 1-norm distance?
A. Euclidean distance
B. Rectilinear (Manhattan) distance
C. The largest dimension value
D. The square root of the sum of squares
Answer: B
Rationale: The 1-norm is the rectilinear (Manhattan) distance .
Q13. What is the infinity norm?
A. Euclidean distance
B. Manhattan distance
C. The value of the largest dimension
D. The sum of all dimensions
Answer: C
Rationale: The infinity norm is the value of the largest dimension .
Q14. In a Support Vector Machine, if the parameter C is set very large, what is most likely to
occur?
