Quiz
Module 3 Review
Create a free account to finish this quiz
Save your progress and see your final score.
Already have an account? Sign in
Questions in this quiz (16)
1F1 score is the ____________ of precision and recall.
- AHarmonic Mean
- BGeometric Mean
- CArithmetic Mean
- DCosine Similarity
2A good metric to evaluate the regression model is:
- AMean Absolute Error
- BConfusion Matrix
- CPrecision
- DRecall
3In regression model if we do not want to penalize the big difference between the Y_Prediction and Y, ________ loss function should be used.
- ARoot Mean Square Log Error
- BRoot Mean Square Error
- CMean Absolute Error
4Which of the following is the method for missing value treatment?
- AImpute the missing values with mean or median
- BColumn Standardization
- CMin-Max Scaling
- DZ-Score Scaling
5______ is a very important principle in data mining projects.
- AGIGO
- BBIGO
- CData Transformation
- DFeature scaling
6_______ can be seen as a measure of quality, and ______ as a measure of quantity.
- APrecision, Recall
- BRecall, Precision
- CPrecision, F1 Score
- DF1 Score, Recall
7Which of the following is not a type of binning transformation?
- ADensity
- BBucket
- COptimal binning for relationship to target
- DQuantile
8When a large number of false positives has a higher cost to business, which machine learning metric is best to be chosen?
- APrecision
- BRecall
- CF1 score
- DSensitivity
9Which of the following is not an imputation method for a missing interval variable?
- ACount
- BMean
- CMedian
- DTree
10Using a confusion matrix, what statistical metric can be calculated with the formula: True Positive/Actual Positive?
- ASensitivity
- BSpecificity
- CMisclassification Rate
- DRecall
11Misclassification Rate
- A(FP+FN)/(FP+FN+TN+TP)
- B(FP*FN)/(FP+FN+TN+TP)
- C(FP-FN)/(FP+FN-TN+TP)
12Sensitivity
- ATP/(FN+TP)
- BTP/(FN-TP)
- CFP/(FN+TP)
- DFN /(FN+TP)
13Actual Negative
- ATN + FN
- BTP + FP
- CTP + FN
- DTN + FP
14AUC measures the entire 2D area under ROC curve from _____
- A(0,0) to (0,1)
- B(-1,0) to (0,0)
- C(0,0) to (1,1)
- D(0,1) to (1,1)
15When a large number of false negatives has a higher cost to business, which machine learning metric is best to be chosen?
- ARecall
- BPrecision
- CAccuracy
- DAUC score
16In the ROC curve, with the increase in False Positive Rate, the True Positive Rate ________
- ADecreases
- BIncreases
- CRemains constant
- DDoesn’t depend on each other
Information
- Description
- Quiz on Module 3 Review, created by Nick Feldman on 01/05/2022.
- Author
- Nick Feldman
- Language
- English
- Views
- 0
- Rating
- No ratings yet Rate this quiz