Week 9: MBA

Question 1

What is “support”?

Answer 1

The fraction of transactions that contain that item/item set

Question 2

What is confidence of a rule (A→B)?

Answer 2

The conditional probability that a transaction that contains the items on the LHS also contain those on the right

Question 3

What is confidence of a rule (A→B)?

Answer 3

The conditional probability that a transaction that contains the items on the LHS also contain those on the right

Question 4

How to calculate confidence?

Answer 4

Support (A->B)/Support A

Question 5

What is lift and how do you calculate it?

Answer 5

It is the ratio by which the confidence of a rule exceeds the expected confidence.

Confidence(A–>B) / Expected confidence (Support B)

Question 6

What is the Apriori principle?

Answer 6

It states that if an itemset is frequent, then all of its subsets must also be frequent

Question 7

What are the steps in the algorithm to generate frequent item sets?

Answer 7

1. Start with all item sets with a single item and compute their support (or count)
2. Remove the ones that do not have minimum support
3. Generate all two-item item sets using the results retained in the previous step and compute their support
4. Remove the ones that do not have minimum support
5. Continue increasing number of items in item sets till you have item sets of all sizes (or desired max size) with the required minimum support (or count)
6. Output: A list of all frequent itemsets in the dataset that meet the support threshold.