Information Technology Reference
In-Depth Information
improve the efficiency greatly. Another is sampling technology, which selects the
data sets for data mining. This method is efficient in the applications which pay
attention to efficiency. The other method is parallel data mining. Because data in
large scale of database are often located in different nodes in network and the
parallel data mining can improve the efficiency apparently. This method is often
used in Web mining on Internet.
(3) Besides of the above, key research topics also contain how to control the total
scale of association rules, how to select and deal with the acquired rules, and how
to discover fuzzy association rules and the mining algorithm with high efficiency.
From the point of view of mining object, it is a valuable problemthat how to
extend rule mining from relation database to text and Web data in the future.
Association rules mine mainly from transaction database, such as sales data in
supermarkets, also called basket data. A transaction typically includes a unique
transaction identity number and a list of the items making up the transaction. In
transaction databases, we will investigate transactions with many items. Suppose
product A appears in transaction 1, product B appears in transaction 2, we want
to know how about product A and B appear in a transaction. In the knowledge
discovery of database, association rule is a knowledge pattern which describes
some products appear meantime in a transaction. Association rule learners are
used to discover elements that co-occur frequently within a data set consisting of
multiple independent selections of elements (such as purchasing transactions),
and to discover rules, such as implication or correlation, which relate
co-occurring elements. Questions such as "if a customer purchases product A,
how likely is he to purchase product B?" and "What products will a customer buy
if he buys products C and D?" are answered by association-finding algorithms.
This application of association rule learners is also known as market basket
analysis. As with most data mining techniques, the task is to reduce a potentially
huge amount of information to a small, understandable set of statistically
supported statements.
Let
R=
{
I
1
,I
2
,
⋯
,I
m
} be a itemset,
W
is a set of transactions. Each transaction
T
which is in W is a itemset,
T
⊆
R
. Suppose
A
is a itemset,
T
is a transaction and
A
⊆
T
, we call transaction
T
supports set
A
.
A
n
association rule is an implication
of the form
A
¼
B
, where
A
,
B
are two itemsets,
A
⊂
R
B
⊂
R
and
A
ŝ
B=
∅.
