Font Size: a A A

The Association Rule Mining Algorithm Design And Implementation,

Posted on:2007-08-18Degree:MasterType:Thesis
Country:ChinaCandidate:X Y CaoFull Text:PDF
GTID:2208360185982501Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
Data mining technology is an effective approach to resolve the problem of abundant data and scanty information.It currently is the research frontier within the information science field.The related researches and applications have greatly improved the ability for decision supporting.It has been deemed to a field that has broad prospect of application in database research.This paper describes the conception,function and patterns of data mining.In many data mining algorithms,mining association rule is an important matter in data mining, in which mining frequent itemsets is a key problem in mining association rule. Because maximum frequent itemsets embrace all frequent itemsets, the problem of mining frequent itemsets is converted to the problem of mining maximum frequent itemsets. Mining maximal frequent itemsets is very important in data mining. This paper studies mostly the problem of mining frequent itemsets and maximal frequent itemsets.The previous algorithm of mining frequent itemsets fall into two classes which contains algorithm basing on Apriori and FP_growth. The algorithm basing on Apriori all need to produce the candidate itemsets and judge whether the candidate itemsets is the frequent itemsets. And the algorithms basing on FP_growth at least skim the database twice in order to constitute according the FP-tree. However it cost largely to delect the candidate itemsets and skim the database. This paper presents mending IODLG Alogorithm which improves the efficiency of mining frequent itemsets as following.Firstly,the algorithm adopts the Bit-project which give every item a bit-value, that can increase the speed of mining frequent itemsets and make the algorithm skim the database only one to get all the information for the mining frequent itemsets. Secondly, the algorithm substitutes item-name for item-value, that can better connect the item bit vector. Thirdly the algorithm imports the out-degree value and in-degree value for every node in the graph,that can efficiently reduce the numbers of candidate itemsets. Through these mending measures, Our experimental...
Keywords/Search Tags:Association rule, Maximum Frequent Itemsets, Frequent Pattern Tree
PDF Full Text Request
Related items