数据挖掘大作业答案

Part I: 书面作业

1. (Data Preprocessing) Data transformation

Normalization: scaled to fall within a small, specified range

1) Problems

Suppose that the data for analysis includes the attribute age. The age values for

the data tuples are(in increasing order): 13,15,16,16,19,20,20,21,

22,22,25,25,25,25,30,33,33,35,35,35,35,36,40,45,46,52,70.

min

󰇛

A

󰇜

= 13, max

󰇛

A

󰇜

= 70

󰇛

󰇜

+ 0.0 = 0.386

󰇛

󰇜

A

󰇜

= 12.94

v

′

=

35 29.96

= 0.389

(c) normalization by decimal scaling

′

10

2

(d) Comment on which method you would prefer to use for the given data, giving

reason as to why.

I prefer to use the normalization by decimal scaling.

For min-max normalization, if new data inserted are out of the range of attribute

age, then “out of bounds” error will occurs.

For z-score normalization, additional calculations and storage have to be taken in

(a)

At the granularity of item_category(eg.item;could be “milk”),for the following

rule template:

List the frequent k-itemset for the largest k and all of the strong association

rules( with their support s and confidence c ) containing the frequent k-itemset for the

At the granularity of brand-item_category(e.g. item; could be “sunset-milk”), for

2) Answers

(a)

Item collection:

󰇝

Crab, Milk, Cheese, Bread,Apple, Pie

, m = 6

Frequent Itemsets( with Apriori Algorithm ):

L =

󰇝

Milk: 4, Cheese: 3, Bread: 4, Milk  Cheese: 3, Milk

 Bread: 4, Cheese  Bread: 3, Milk  Cheese  Bread: 3

󰇞

support(Milk)=4/4=100%,support(Bread)=4/4=100%,

K=3

L =

󰇝

Milk  Cheese  Bread: 3

Association rules:

󰇛

󰇜

= P

󰇛

A



B

󰇜

=

P(A  B)

P(A)

=

Strong association rules( min_conf=80% ):