DSCI 2010
Data Science Essentials
Final Exam (Qns & Ans)
2025
General Instructions
1. Read All Questions Carefully: Make sure you understand each question.
2. Time Management: You have a specific amount of time to complete the exam.
Keep an eye on the clock and pace yourself.
3. Allowed Materials: Only use materials that are explicitly allowed. Unauthorized
materials can lead to disqualification.
4. ANS Format: Follow the required format for your ANS. For example, multiple-
choice questions might need you to select the best ANS, while essay questions
require detailed responses.
5. Academic Integrity: Adhere to the university's honor code. Any form of cheating or
plagiarism is strictly prohibited.
6. Technical Requirements: Ensure your computer and internet connection are
stable. For online exams, you might need a webcam and microphone for proctoring
purposes.
7. Submission: Submit your ANS before the time expires. Late submissions might
not be accepted.
©2024/2025
,1. What is underfitting in a machine learning model?
a) The model is too complex
b) The model is too simple
c) The model performs well on training data but poorly on
unseen data
d) None of the above
ANS: b) The model is too simple
Rationale : Underfitting occurs when a model is too simplistic
and fails to capture the underlying patterns in the data.
2. Which of the following can be considered an example of
supervised learning?
a) Clustering
b) Regression
c) Anomaly Detection
d) Dimensionality Reduction
ANS: b) Regression
Rationale : Supervised learning uses labeled data to learn, of
which regression tasks predict a continuous outcome.
©2024/2025
, 3. In the context of data normalization, what is min-max scaling?
a) Transforming features to have mean=0 and variance=1
b) Rescaling the features to a fixed range of [0, 1]
c) Removing outliers from the dataset
d) Summarizing data using means and medians
ANS: b) Rescaling the features to a fixed range of [0, 1]
Rationale : Min-max scaling adjusts the feature values to a
specified range, typically between 0 and 1.
4. Which algorithm would be most appropriate for predicting
categorical outcomes?
a) Linear Regression
b) Decision Trees
c) K-Means Clustering
d) Principal Component Analysis
ANS: b) Decision Trees
Rationale : Decision trees can model categorical outcomes
effectively, making them suitable for classification tasks.
5. What does the 'Curse of Dimensionality' refer to?
©2024/2025
Data Science Essentials
Final Exam (Qns & Ans)
2025
General Instructions
1. Read All Questions Carefully: Make sure you understand each question.
2. Time Management: You have a specific amount of time to complete the exam.
Keep an eye on the clock and pace yourself.
3. Allowed Materials: Only use materials that are explicitly allowed. Unauthorized
materials can lead to disqualification.
4. ANS Format: Follow the required format for your ANS. For example, multiple-
choice questions might need you to select the best ANS, while essay questions
require detailed responses.
5. Academic Integrity: Adhere to the university's honor code. Any form of cheating or
plagiarism is strictly prohibited.
6. Technical Requirements: Ensure your computer and internet connection are
stable. For online exams, you might need a webcam and microphone for proctoring
purposes.
7. Submission: Submit your ANS before the time expires. Late submissions might
not be accepted.
©2024/2025
,1. What is underfitting in a machine learning model?
a) The model is too complex
b) The model is too simple
c) The model performs well on training data but poorly on
unseen data
d) None of the above
ANS: b) The model is too simple
Rationale : Underfitting occurs when a model is too simplistic
and fails to capture the underlying patterns in the data.
2. Which of the following can be considered an example of
supervised learning?
a) Clustering
b) Regression
c) Anomaly Detection
d) Dimensionality Reduction
ANS: b) Regression
Rationale : Supervised learning uses labeled data to learn, of
which regression tasks predict a continuous outcome.
©2024/2025
, 3. In the context of data normalization, what is min-max scaling?
a) Transforming features to have mean=0 and variance=1
b) Rescaling the features to a fixed range of [0, 1]
c) Removing outliers from the dataset
d) Summarizing data using means and medians
ANS: b) Rescaling the features to a fixed range of [0, 1]
Rationale : Min-max scaling adjusts the feature values to a
specified range, typically between 0 and 1.
4. Which algorithm would be most appropriate for predicting
categorical outcomes?
a) Linear Regression
b) Decision Trees
c) K-Means Clustering
d) Principal Component Analysis
ANS: b) Decision Trees
Rationale : Decision trees can model categorical outcomes
effectively, making them suitable for classification tasks.
5. What does the 'Curse of Dimensionality' refer to?
©2024/2025
