AN 300 -EXAM 2 WITH QUESTIONS AND
ANSWERS
Which of the following is true about cluster analysis? Check all that
apply.
-It is a descriptive analytics technique
-It is to discover associations between objects.
-It is used to discover natural groupings of objects.
-It is to answer what has happened questions.
-It is a descriptive analytics method.
-It is to answer what could happen questions. - ANSWER -It is a
descriptive analytics technique.
-It is used to discover natural groupings of objects.
-It is to answer what has happened questions.
Which of the following is a characteristic of a cluster analysis
problem? (Check all that apply):
-The data that describes the object must be given.
-Its objective is to maximize similarities of objects between groups.
-The data on group memberships must be given.
-It is about how to discover
-It is about how to organize objects into groups.
-Its objective is to maximize similarities of objects within groups. -
ANSWER -The data that describes the object must be given.
-It is about how to organize objects into groups.
,-Its objective is to maximize similarities of objects within groups.
You have data on the weight and height of patients. Which similarity
measure should be used to calculate how similar a group of patients
is to one another?
-Manhattan coefficient
-Correlation coefficient
-Euclidean distance
-Matching coefficient
-Straight line distance - ANSWER -Euclidean distance
You have data on the gender and income levels of customers. Which
similarity measure should be used to calculate how similar a group of
customers is to one another?
-Correlation coefficient
-Matching coefficient
-Straight line distance
-Manhattan coefficient
-Euclidean distance - ANSWER -Matching coefficient
Match the description on the left with the measure to use on the right.
-It is the length of a straight line between two objects. - ANSWER
Euclidian Distance
It requires replacing the raw value of data with its z-score. -
ANSWER Euclidian Distance
The lower the measure the better - ANSWER Euclidian Distance
It is for categorical Data - ANSWER Matching Coefficient
It is a ratio of number of columns with matching categorical values to
the total number of categorical columns. - ANSWER Matching
, Coefficient
Which of the following is a business application of cluster analysis?
-Affinity analysis
-Outlier detection
-Market basket analysis
-Sentiment analysis - ANSWER -Outlier detection
Match the situation on the left with the clustering method on the right.
-You have 300 rows of data. - ANSWER Hierarchical Clustering
You only have numerical data to work with. - ANSWER K-Means
Clustering
You have outliers in the data. - ANSWER K-Means Clustering
You want to experiment with different ways to calculate the distance
between clusters. - ANSWER Hierarchical Clustering
Which of the following is true about CCC? Check all that apply.
-A value greater than two is highly desirable.
-It is used to tell the goodness of fit of clustering.
-It stands for Cubic Clustering Criterion.
-It is related to the proportion of variance in the data accounted for
by the clusters.
-It is related to the proportion of matching values in a column against
all columns.
-A negative value is impossible.
-It stands for Complete Clustering Criterion. - ANSWER -A value
greater than two is highly desirable.
-It is used to tell the goodness of fit of clustering.
ANSWERS
Which of the following is true about cluster analysis? Check all that
apply.
-It is a descriptive analytics technique
-It is to discover associations between objects.
-It is used to discover natural groupings of objects.
-It is to answer what has happened questions.
-It is a descriptive analytics method.
-It is to answer what could happen questions. - ANSWER -It is a
descriptive analytics technique.
-It is used to discover natural groupings of objects.
-It is to answer what has happened questions.
Which of the following is a characteristic of a cluster analysis
problem? (Check all that apply):
-The data that describes the object must be given.
-Its objective is to maximize similarities of objects between groups.
-The data on group memberships must be given.
-It is about how to discover
-It is about how to organize objects into groups.
-Its objective is to maximize similarities of objects within groups. -
ANSWER -The data that describes the object must be given.
-It is about how to organize objects into groups.
,-Its objective is to maximize similarities of objects within groups.
You have data on the weight and height of patients. Which similarity
measure should be used to calculate how similar a group of patients
is to one another?
-Manhattan coefficient
-Correlation coefficient
-Euclidean distance
-Matching coefficient
-Straight line distance - ANSWER -Euclidean distance
You have data on the gender and income levels of customers. Which
similarity measure should be used to calculate how similar a group of
customers is to one another?
-Correlation coefficient
-Matching coefficient
-Straight line distance
-Manhattan coefficient
-Euclidean distance - ANSWER -Matching coefficient
Match the description on the left with the measure to use on the right.
-It is the length of a straight line between two objects. - ANSWER
Euclidian Distance
It requires replacing the raw value of data with its z-score. -
ANSWER Euclidian Distance
The lower the measure the better - ANSWER Euclidian Distance
It is for categorical Data - ANSWER Matching Coefficient
It is a ratio of number of columns with matching categorical values to
the total number of categorical columns. - ANSWER Matching
, Coefficient
Which of the following is a business application of cluster analysis?
-Affinity analysis
-Outlier detection
-Market basket analysis
-Sentiment analysis - ANSWER -Outlier detection
Match the situation on the left with the clustering method on the right.
-You have 300 rows of data. - ANSWER Hierarchical Clustering
You only have numerical data to work with. - ANSWER K-Means
Clustering
You have outliers in the data. - ANSWER K-Means Clustering
You want to experiment with different ways to calculate the distance
between clusters. - ANSWER Hierarchical Clustering
Which of the following is true about CCC? Check all that apply.
-A value greater than two is highly desirable.
-It is used to tell the goodness of fit of clustering.
-It stands for Cubic Clustering Criterion.
-It is related to the proportion of variance in the data accounted for
by the clusters.
-It is related to the proportion of matching values in a column against
all columns.
-A negative value is impossible.
-It stands for Complete Clustering Criterion. - ANSWER -A value
greater than two is highly desirable.
-It is used to tell the goodness of fit of clustering.
