2026-09-21T23:03:30Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4590 2022-03-24T23:16:21Z 1582963302567:1597824273898 user-ucsy

Frequent Pattern Mining for Stream Data by Using Hadoop GM-Tree and GTree Aung, Than Htike Kham, Nang Saing Moon Since origination of mining, frequent patternmining has become a mandatory issue in datamining. Transaction process for mining pattern needsefficient data structures and algorithms. This systemproposed tree structure, called GMTree(Generateand Merge Tree)-GTree(Group Tree), which is ahybrid of prefix based incremental mining usingcanonical order tree and batch incrementingtechniques. Proposed system make the tree structuremore compact, canonically ordered of nodes andavoids sequential incrementing of transactions. Itgives a scalable algorithm with minimum overheadsof modifying the tree structure during updateoperations. It operates on extremely largetransaction database in dynamic environment whichis especially expected to give better results in thiscase.The proposed system used Apache Hadoop andhybrid GMTree-GTree. The results shows Hadoopimplementation of algorithm performs more timesbetter than in Java. 2018-02-22 http://hdl.handle.net/20.500.12678/0000004590 https://meral.edu.mm/records/4590