Suppose a store sells 5 products A, B, C, D, and E. Below are 11 baskets of products purchased at the store.
Basket Items: 0 DEA, 1 A B, 2 BCAE, 3 DC, 4 CBEA, 5 DEB, 6 BAC, 7 CDEB, 8 BAD, 9 ECB, 10 CAE
R = 8
Before you work on the questions below, DELETE Basket R from the dataset. E.g. if your R is 5, then you use basket 0-4 and 6-10 to answer the questions.
a) Find the support of {A}, {B}, {C}, {A,B}, {A,C}, {B,C}, and {A,B,C} (2 marks)
(b) Find the support, confidence, and lift of the association rule {A,B} => {C}. Correct your answers to 4 decimal places. (5 marks)
(c) Explain whether {A,B} and {C} are positively associated, negatively associated, or independent. (2 marks)
(d) In another set of data, {A,B,C}, {B,C,D}, {B,C,E}, {B,D,E} are 3-itemsets with support greater than found by the Apriori algorithm. What is/are the candidates of 4-itemsets that will be produced by the self-joining method? (1 mark)
Solution:
(a)
Itemset
{A}
{B}
{C}
{A,B}
{A,C}
{B,C}
{A,B,C}
(a)
Support
2 marks
(b) Support:
(b)
5 marks
Confidence:
Lift: