Data Science

caret

Question 1
Marks : +2 | -2
Pass Ratio : 100%
Which of the following function can be used to flag predictors for removal?
searchCorrelation
findCausation
findCorrelation
none of the mentioned
Explanation:
Some models thrive on correlated predictors.
Question 2
Marks : +2 | -2
Pass Ratio : 100%
The function preProcess estimates the required parameters for each operation.
True
False
Explanation:
predict.preProcess is used to apply them to specific data sets.
Question 3
Marks : +2 | -2
Pass Ratio : 100%
Point out the wrong statement.
The trapezoidal rule is used to compute the area under the ROC curve
For regression, the relationship between each predictor and the outcome is evaluated
An argument, para, is used to pick the model fitting technique
All of the mentioned
Explanation:
An argument, nonpara, is used to pick the model fitting technique.
Question 4
Marks : +2 | -2
Pass Ratio : 100%
Point out the correct statement.
The difference between the class centroids and the overall centroid is used to measure the variable influence
The Bagged Trees output contains variable usage statistics
Boosted Trees uses different approach as a single tree
None of the mentioned
Explanation:
The larger the difference between the class centroid and the overall center of the data, the larger the separation between the classes.
Question 5
Marks : +2 | -2
Pass Ratio : 100%
varImp is a wrapper around the evimp function in the _______ package.
numpy
earth
plot
none of the mentioned
Explanation:
The earth package is an implementation of Jerome Friedman’s Multivariate Adaptive Regression Splines.
Question 6
Marks : +2 | -2
Pass Ratio : 100%
For most classification models, each predictor will have a separate variable importance for each class.
True
False
Explanation:
The exceptions are classification trees, bagged trees and boosted trees.
Question 7
Marks : +2 | -2
Pass Ratio : 100%
Which of the following model sums the importance over each boosting iteration?
Boosted trees
Bagged trees
Partial least squares
None of the mentioned
Explanation:
gbm package can be used here.
Question 8
Marks : +2 | -2
Pass Ratio : 100%
The advantage of using a model-based approach is that is more closely tied to the model performance.
True
False
Explanation:
Model-based approach is able to incorporate the correlation structure between the predictors into the importance calculation.
Question 9
Marks : +2 | -2
Pass Ratio : 100%
The preProcess class can be used for many operations on predictors.
True
False
Explanation:
Operations include centering and scaling.
Question 10
Marks : +2 | -2
Pass Ratio : 100%
Which of the following can be used to generate balanced cross–validation groupings from a set of data?
createFolds
createSample
createResample
none of the mentioned
Explanation:
createResample can be used to make simple bootstrap samples.