The Performance Improvements of SPRINT Algorithm Based on the Hadoop Platform

摘要

The emergence of cloud computing provide the medium enterprises many low-cost mass data analysis solutions. Decision tree algorithm in which one of the biggest problems is its computational complexity is proportional to the size and training data, resulting in a large number of computing time in constructing Data Set. The article aim at the SPRINT algorithm based on the Hadoop platform, presenting a parallel method of constructing a decision tree and then solving the parallel problem in Hadoop platform.