A Survey of Efficient Algorithms and New Approach for Fast Discovery of Frequent Itemset for Association Rule Mining (DFIARM)

Abstract

[15] is a popular data mining technique because of its wide application in marketing and retail communities as well as other more diverse fields [18]. Association rule mining is a method of finding relationships of the form X→Y amongst Item-sets that occur together in a database where X and Y are disjoint Item-sets [17]. Support and confidence measures serve as the basis for customary techniques in association rule mining. The support and confidence are predefined by users to drop the rules that are not so interesting or useful. The association rule indicates that the transactions that contain X tend to also contain Y. Suppose the support of an item is 0.1%, it means only 0.1 percent of the transaction contain purchasing of this item [16]. The task of mining association rules is defined as follows: Let IS ={i1,i2,i3,…,im} a set of items and TDI ={t1,t2,t3, …,tn}be a set of transaction data items, where ti = {ISi1, ISi2, ISi3, ……, ISip}, P≤ m and ISij ∈ IS, if X ⊆ I with k=|X| is called a k-item-set or simply an item-set. An An expression, where X, Y are item-sets and association rule X X∩Y= Φ holds is called an association rule X → Y. The measure of number of transactions T supporting an item set X with respect to TDI is termed as the Support of an item-set. Support (X) = | {T ∈ TDI| X⊆ T}| TDI| ………….(1) The ratio of the number of transactions that hold X U Y to the number of transactions that holds X is said to be the confidence of an association rule X → Y. Conf (X → Y)=Support (X U Y) / Support (X) ………….(2) Informally, the prediction using association rule set can be described as follows. For a given association rule set R and an Item-set P, we say that the predictions for P from R is a sequence of items Q. The sequence of Q is generated by using the rules in R which is descending order of confidence. For each rule r that matches P (i.e. for each rule whose antecedent is a subset of P), each consequent of r is added to Q. After adding a consequence to Q, all rules whose consequences are in Q are removed form R. The following example shows the association rules of a simple data set and its application to prediction. Example 1: Consider a small database shown in table-1. For minimum support 0.5 and minimum confidence 0.5. For the rule: a b, the 67% is called the support of the rule is the percentage of transactions that contain both a and b. The 80% here called the confidence of the rule, which means that 80% of transaction that contains X also contains Y. Therefore, set of 12 association rules can be found, as shown in Table-2.