A COMPARATIVE STUDY OF DATA PARTITIONING IN FREQUENT ITEMSET MINING ON HADOOP MAPREDUCE CLUSTER

Main Article Content

Jitha Janardhanan
Dr .E.Mary Shyla

Abstract

Distributed parallel algorithms for mining frequent balanced itemsets aims to load by equally dividingdata among a collectionof computing nodes. Over the history, frequent itemsets basedparallel algorithm methods have been illustrated in the literature. In this comparative study aims to present a study of Frequent pattern miningtechniques deviations among in Hadoop MapReduceconcept under the data mining techniques that are in use in large databasetransactions broadcasted among computing nodes. Number of comparative studies has been performed to assess the performance of MapReduce cases and the outcome discloses that Spark Frameworkwith advanced load balancing strategy having better performance than other predictive methods like Apriori, Randomized algorithms.

Downloads

Download data is not yet available.

Article Details

Section
Articles