一种改进的SPRINT算法

An Improved SPRINT Algorithm

自大数据时代以来,数据密集型计算已经引起了相当大的关注.数据密集型计算环境中的数据挖掘研究仍处于初级阶段.提出一种基于MapReduce编程框架和SPRINT算法的决策树分类算法M-BCBT. M-BCBT继承了MapReduce的优点,使算法更适合数据密集型计算应用.算法的性能根据实例进行分析评估.实验结果表明,MBCBT可以缩短操作时间,提高大数据环境的准确性.

Data-intensive computing has attracted considerable attention since the advent of the big data era. Data mining research in data-intensive computing environments is still in its infancy. This paper proposes a decision tree classification algorithm M-BCBT based on MapReduee programming framework and SPRINT algorithm. M-BCBT inherits the advantages of MapReduee, making the algorithm more suitable for data-intensive computing applications. The performance of the algorithm is evaluated based on living examples. Experimental results show that M-BCBT can shorten the operation time and improve the accuracy of big data environment.