Chapter 9: Unsupervised Machine Learning

Question 1

What is a popular algorithm for identifying clusters?

Answer 1

k-means

Question 2

What is k-means?

Answer 2

K in k-means represents the number of clusters, or groupings.
Note that k-means works only for numerical data; that is, all attributes being considered have numerical values.

Question 3

K-means is acceptable to use for master data?

Answer 3

incorrect, only transactional data

Question 4

What can be used for clustering categorical data?

Answer 4

k-modes

Question 5

What is the optimum number of clusters?

Answer 5

between 1 and N

Question 6

What is cluster size?

Answer 6

the number of members within a cluster

Question 7

What is cluster density?

Answer 7

more customers are in a cluster versus another cluster

Question 8

What is cluster distance?

Answer 8

Indicates how dissimilar customers in one cluster are from customers in another cluster.

Question 9

What is an association analysis or “affinity analysis?

Answer 9

A type of unsupervised descriptive data model, it is used to find the hidden connections between sets of items the frequently occur together.

Question 10

“{PIZZA-BY-THE-SLICE} → {SOFT DRINK 20 OZ}” is an example of what?

Answer 10

association rules or associational analysis

Question 11

Businesses can use these association rules to do what?

Answer 11

To promote and recommend items that often occur together.

Question 12

What is an example of a {antecedent(s)} → {consequent(s)}?

Answer 12

{Knee Pads} → {Off Road Helmet}

{Deluxe Touring Bike Black, Elbow Pads} → {Off Road Helmet}

Question 13

What is the support rule in association rules?

Answer 13

Support for a rule is the fraction or percentage of transactions that contain all of the items within the rule

Question 14

Excessive numbers of rules represent an ______ to effective analysis

Answer 14

obstacle

Question 15

Confidence is described as?

Answer 15

Confidence is the measure or probability of the consequent items in transactions that contain the antecedent items. It is a conditional probability

Question 16

What is used to measure the absence of the antecedent?

Answer 16

a Lift

Question 17

What is a Lift?

Answer 17

Lift is the measure of how accurately a rule depicts affinity or association compared to the random (coincidental) co-occurrence of the items.

Question 18

Lift = _______________________?

Answer 18

Lift = confidence of a rule / support of the consequent

Question 19

Lift values greater than _______ imply that the antecedent and consequent are associated (correlated) with each other..

Answer 19

1

Question 20

Lift values less than 1 imply what?

Answer 20

The two items are negatively correlated.

Question 21

What does a apriori algorithm do?

Answer 21

Makes the association analysis straightforward and trims out the infrequent rules by default.