一种基于SPRINT的判定树分类算法

A classification algorithm of decision tree based on SPRINT

分类算法的研究一直是数据挖掘领域的一个热点。本文在介绍了目前判定树分类算法状况的基础上,着重分析了目前较为流行的SPRINT算法,并对该算法的数据结构进行了改进,提出一种新的算法NS,使得分类的速度及可扩充性得以提高。这一改进算法在一个银行决策支持系统(DSS)中得到实现,取得了很好的效果。

Classification is an important problem in the emerging field of data mining. Based on the introduction to some typical classification algorithms, this paper places emphasis on the algorithm of SPRINT which is popular classifier, and presents a new algorithm named NS which includes some improvements based on SPRINT's data structure. The NS makes classification more scalable.