Frequent Pattern Mining Using Modified CP-Tree for Knowledge Discovery
Mining frequent pattern from databases is useful for knowledge discovery. In this paper, we propose modified CP-Tree, which scans entire transactions only once and constructs the tree by inserting the transactions one by one. The constructed tree consists of an item list along with its occurrence. In addition, a sorted order of items with its frequency of occurrence is maintained and based on the sorted value, the tree is dynamically rearranged. In rearranging phase, the nodes are rearranged in each branch based on sorted order of items. Each path of the branch is removed from the tree, sorted based on sorted order of items and inserted back as a branch into the tree. We have evaluated the performance of the proposed modified tree on benchmark databases such as CHESS, MUSHROOM and T10I4D100K. It is observed that the time taken for extracting frequent item from the tree is encouraging compared to conventional CP-Tree.
Frequent Pattern Mining Modified CP-Tree Knowledge Discovery
R .Vishnu Priya A.Vadivel R.S. Thakur
Department of Computer Applications National Institute of Technology Tamilnadu India Maulana Azad National Institute of Technology Madyapradesh India
国际会议
6th International Conference on Advanced Data Mining and Applications(第六届先进数据挖掘及应用国际会议 ADMA 2010)
重庆
英文
254-261
2010-11-19（万方平台首次上网日期，不代表论文的发表时间）