Distributed Communication Decision Tree Algorithm for Disseminated and Heterogeneous Environment

Abstract

Current Classification algorithms require large amounts of data to be stored enduringly in the memory for long assortment and amount of time. Diverse classification techniques had been already proposed in the literature for both in the run of the mill environment and distributed environment. Mining of decision trees in the distributed environment can be able to handle the large amount of data but with high communication cost. A new distributed communication decision tree algorithm is proposed here which reduces the communication cost for the transmission of the data in the distributed and heterogeneous environment.