Download now Free registration required
In this paper, the authors present a new algorithm for mining generalized association rules. They develop the algorithm which scans database one time only and use Tidset to compute the support of generalized itemset faster. A tree structure called GIT-tree, an extension of IT-tree, is developed to store database for mining frequent itemsets from hierarchical database. Their algorithm is often faster than MMS Cumulate, an algorithm mining frequent itemsets in hierarchical database with multiple minimum supports, in experimental databases. Mining association rules plays an important role in knowledge discovery and data mining (KDD). Its purpose is mining the hidden knowledge in databases.
- Format: PDF
- Size: 176.7 KB