摘要
本文对利用数据挖掘技术解决大数据量的处理问题进行了较深入的研究,基于粗糙集理论提出了基于知识依赖的属性约简方法。该方法能够快速有效地找出真正影响决策的属性。此外,又提出了以知识依赖为启发规则的决策树生成方法。该方法预测精度高,计算简便。
This paper studies the methods of processing massive data with data mining techniques, and proposes an attribute reduction method based on knowledge dependency with the rough set theory, which can quickly and effectively find the attributes that really affect decisions. In addition, the paper puts forth an algorithm for generating decision trees based on knowledge dependency,which features high prediction precision and easy computation.
出处
《计算机工程与科学》
CSCD
2008年第7期108-109,113,共3页
Computer Engineering & Science
关键词
决策树
生成算法
知识依赖
decision tree
generation algorithm
knowledge dependency