D491 DTAN 3100
Introduction to Analytics
2.0 Credits
Objective Assessment Review (Qns &
Ans)
2025
©2025
, Question 1 (Multiple Choice):
_A leading technology firm is analyzing a high-dimensional customer dataset that
suffers from multicollinearity. Which algorithm is most appropriate to handle this
problem while ensuring interpretability?_
A. K-Nearest Neighbors (KNN)
B. Ridge Regression
C. Decision Trees
D. Support Vector Machines (SVM)
Correct ANS: B. Ridge Regression
Rationale:
Ridge Regression applies L2 regularization, which helps to reduce the effect of
multicollinearity by shrinking the coefficients. This ensures that the model remains
interpretable and avoids overfitting in high-dimensional settings.
---
Question 2 (Multiple Choice):
_In preparing the data for an analytics project, an analyst needs to handle features
with different scales. Which technique is the most suitable for this scenario?_
A. Data smoothing
B. Feature extraction
C. Feature scaling (Normalization/Standardization)
D. Dimensionality reduction
Correct ANS: C. Feature scaling (Normalization/Standardization)
Rationale:
Normalization and standardization adjust the range of feature values, ensuring that
each feature contributes equally to the model and that algorithms sensitive to scale
perform optimally.
---
Question 3 (Multiple Choice):
©2025
, _In a predictive modeling case study, a company wishes to forecast sales. The data
scientist is choosing between multiple regression models. Which criterion is
essential when selecting the final model?_
A. The model with the highest training accuracy
B. The model with the lowest computational cost
C. The model that best balances bias and variance based on cross-validation
D. The model with the most complex architecture
Correct ANS: C. The model that best balances bias and variance based on cross-
validation
Rationale:
In predictive modeling, especially when forecasting sales, balancing bias
(underfitting) and variance (overfitting) is crucial. Cross-validation helps ensure that
the model generalizes well on unseen data.
---
Question 4 (Multiple Choice):
_A data analytics team is evaluating the impact of external factors on website
traffic. Which approach is best suited to uncover and control for confounding
factors in their analysis?_
A. Simple linear regression
B. Time series analysis without any adjustment
C. Multivariate regression incorporating control variables
D. Clustering analysis
Correct ANS: C. Multivariate regression incorporating control variables
Rationale:
Multivariate regression allows for the inclusion of multiple independent variables
and control variables, which helps isolate the effect of each factor and account for
potential confounders.
---
Question 5 (Multiple Choice):
_Consider a case where a firm must process massive volumes of data generated in
©2025
Introduction to Analytics
2.0 Credits
Objective Assessment Review (Qns &
Ans)
2025
©2025
, Question 1 (Multiple Choice):
_A leading technology firm is analyzing a high-dimensional customer dataset that
suffers from multicollinearity. Which algorithm is most appropriate to handle this
problem while ensuring interpretability?_
A. K-Nearest Neighbors (KNN)
B. Ridge Regression
C. Decision Trees
D. Support Vector Machines (SVM)
Correct ANS: B. Ridge Regression
Rationale:
Ridge Regression applies L2 regularization, which helps to reduce the effect of
multicollinearity by shrinking the coefficients. This ensures that the model remains
interpretable and avoids overfitting in high-dimensional settings.
---
Question 2 (Multiple Choice):
_In preparing the data for an analytics project, an analyst needs to handle features
with different scales. Which technique is the most suitable for this scenario?_
A. Data smoothing
B. Feature extraction
C. Feature scaling (Normalization/Standardization)
D. Dimensionality reduction
Correct ANS: C. Feature scaling (Normalization/Standardization)
Rationale:
Normalization and standardization adjust the range of feature values, ensuring that
each feature contributes equally to the model and that algorithms sensitive to scale
perform optimally.
---
Question 3 (Multiple Choice):
©2025
, _In a predictive modeling case study, a company wishes to forecast sales. The data
scientist is choosing between multiple regression models. Which criterion is
essential when selecting the final model?_
A. The model with the highest training accuracy
B. The model with the lowest computational cost
C. The model that best balances bias and variance based on cross-validation
D. The model with the most complex architecture
Correct ANS: C. The model that best balances bias and variance based on cross-
validation
Rationale:
In predictive modeling, especially when forecasting sales, balancing bias
(underfitting) and variance (overfitting) is crucial. Cross-validation helps ensure that
the model generalizes well on unseen data.
---
Question 4 (Multiple Choice):
_A data analytics team is evaluating the impact of external factors on website
traffic. Which approach is best suited to uncover and control for confounding
factors in their analysis?_
A. Simple linear regression
B. Time series analysis without any adjustment
C. Multivariate regression incorporating control variables
D. Clustering analysis
Correct ANS: C. Multivariate regression incorporating control variables
Rationale:
Multivariate regression allows for the inclusion of multiple independent variables
and control variables, which helps isolate the effect of each factor and account for
potential confounders.
---
Question 5 (Multiple Choice):
_Consider a case where a firm must process massive volumes of data generated in
©2025
