DATA MINING EXAM #2 QUESTIONS
WITH COMPLETE SOLUTIONS
Principal components analysis (PCA) is a useful procedure for reducing the number
of predictors in the model by analyzing the....
A. Output Variables
B. Input Variables
C. Input and Output Variables
D. Categorical Variables - Answer-Answer: B - Input Variables (Chapter 4, pg. 78 )
True or False: Pivot tables can only be used for one variable. - Answer-Answer:
False: Pivot tables can be used for multiple variables. For categorical variables we
obtain a breakdown of the records by the combination of categories.
Chapter 4, page 75, Pivot tables
________ is a way to reduce the number of predictors in a model by analyzing input
variables.
Pivot Tables
Principle Component Analysis
Correlation Analysis
Dimension Reduction - Answer-Answer: b. Principle Component Analysis; Principle
Component Analysis is a way to reduce the number of predictors in a model by
analyzing input variables. It is especially useful when we have highly correlated
subsets of measurements.
Chapter 4, page 79, Principle Component Analysis
There are 4 steps in the Principal Component Analysis - Answer-T - 1. The
correlation matrix for all varaibles is computed. 2. Factor extraction. 3. Factor rotation
4. Make final decisions about the number of underlying factors
The definitions of Sensitivity and Specificity are:
A. Sensitivity = TP/(TP+FN) Specificity = TN/(TN+FP)
B. Sensitivity = TN/(TN+FP) Specificity = TP/(TP+FN)
C. Sensitivity = TN/(TP+FN) Specificity = TP/(TN+FP)
D. Sensitivity = TP/(TN+FN) Specificity = TN/(TP+FP) - Answer-Solution: A
1)___________ gives an idea of systematic over- or under-prediction
, A. MAD (Mean absolute deviation)
B. Average error
C. RMSE (Root-mean-squared-error
D. MAPE (mean absolute percentage error) - Answer-Answer: B. Average Error
Source Chapter 5 power point slide 45
2)True or False: An important consideration in selecting a forecasting method is the
accuracy of the forecast. - Answer-Answer: TRUE
Source Chapter 5 powerpoint slide 46
Which of the following is an alternative for plotting large amounts of data that display
each individual observation?
A. Breaking down the data into subsets
B. Sampling
C. Using jittering
D. All of the above. - Answer-Answer: D. All of the above.
Counts & percentages are useful for summarizing categorical data. - Answer-
Answer: True
Rationale: Chapter 4 power point Slides
True or False. In principal component analysis a boxplot analysis is used to find and
describe the underlying principal components driving data values for a large set of
variables. - Answer-FALSE. It's a correlational method that is used(Chapter 4
powerpoint)
Which of the following is a good alternative for plotting a large number of data that
displays each individual observation?
A. Breaking down the data into subsets
B. Sampling
C. Using jittering
D. All of the above. - Answer-D. All of the above. (Chapter 3 page58)
Which of the following is not a quantitative graphical method? A. Histogram B. Bar
Chart C. Scatter Diagram D. Box Plot E. Stem-and-Leaf Diagram - Answer-Answer:
B. box chart (chapter 4)
rue or False.. A histogram is a tabular method that is used on Quantitative data. -
Answer-Answer : False. Histograms are used on quantitative data but it is a
graphical method. (Slide 3 in Chapter 3)
All of the following are steps in Factor Analysis except for..
A. Factor extraction
WITH COMPLETE SOLUTIONS
Principal components analysis (PCA) is a useful procedure for reducing the number
of predictors in the model by analyzing the....
A. Output Variables
B. Input Variables
C. Input and Output Variables
D. Categorical Variables - Answer-Answer: B - Input Variables (Chapter 4, pg. 78 )
True or False: Pivot tables can only be used for one variable. - Answer-Answer:
False: Pivot tables can be used for multiple variables. For categorical variables we
obtain a breakdown of the records by the combination of categories.
Chapter 4, page 75, Pivot tables
________ is a way to reduce the number of predictors in a model by analyzing input
variables.
Pivot Tables
Principle Component Analysis
Correlation Analysis
Dimension Reduction - Answer-Answer: b. Principle Component Analysis; Principle
Component Analysis is a way to reduce the number of predictors in a model by
analyzing input variables. It is especially useful when we have highly correlated
subsets of measurements.
Chapter 4, page 79, Principle Component Analysis
There are 4 steps in the Principal Component Analysis - Answer-T - 1. The
correlation matrix for all varaibles is computed. 2. Factor extraction. 3. Factor rotation
4. Make final decisions about the number of underlying factors
The definitions of Sensitivity and Specificity are:
A. Sensitivity = TP/(TP+FN) Specificity = TN/(TN+FP)
B. Sensitivity = TN/(TN+FP) Specificity = TP/(TP+FN)
C. Sensitivity = TN/(TP+FN) Specificity = TP/(TN+FP)
D. Sensitivity = TP/(TN+FN) Specificity = TN/(TP+FP) - Answer-Solution: A
1)___________ gives an idea of systematic over- or under-prediction
, A. MAD (Mean absolute deviation)
B. Average error
C. RMSE (Root-mean-squared-error
D. MAPE (mean absolute percentage error) - Answer-Answer: B. Average Error
Source Chapter 5 power point slide 45
2)True or False: An important consideration in selecting a forecasting method is the
accuracy of the forecast. - Answer-Answer: TRUE
Source Chapter 5 powerpoint slide 46
Which of the following is an alternative for plotting large amounts of data that display
each individual observation?
A. Breaking down the data into subsets
B. Sampling
C. Using jittering
D. All of the above. - Answer-Answer: D. All of the above.
Counts & percentages are useful for summarizing categorical data. - Answer-
Answer: True
Rationale: Chapter 4 power point Slides
True or False. In principal component analysis a boxplot analysis is used to find and
describe the underlying principal components driving data values for a large set of
variables. - Answer-FALSE. It's a correlational method that is used(Chapter 4
powerpoint)
Which of the following is a good alternative for plotting a large number of data that
displays each individual observation?
A. Breaking down the data into subsets
B. Sampling
C. Using jittering
D. All of the above. - Answer-D. All of the above. (Chapter 3 page58)
Which of the following is not a quantitative graphical method? A. Histogram B. Bar
Chart C. Scatter Diagram D. Box Plot E. Stem-and-Leaf Diagram - Answer-Answer:
B. box chart (chapter 4)
rue or False.. A histogram is a tabular method that is used on Quantitative data. -
Answer-Answer : False. Histograms are used on quantitative data but it is a
graphical method. (Slide 3 in Chapter 3)
All of the following are steps in Factor Analysis except for..
A. Factor extraction
