文献类型：期刊文献

中文题名：基于RBF网络的粗糙集属性约简策略

英文题名：Reduction of Rough Sets Attributes Based on RNF Networks

作者：崔明义[1];耿向平[1]

第一作者：崔明义

机构：[1]河南财经学院计算机科学系

第一机构：河南财经政法大学计算机与信息工程学院

年份：2006

期号：3

起止页码：23-26

中文期刊名：安阳工学院学报

外文期刊名：Journal of Anyang Institute of Technology

收录：国家哲学社会科学学术期刊数据库

基金：河南省自然科学基金资助项目(0411013800;0411014500);河南省高校杰出科研人才创新工程项目(2004KYCX014)

语种：中文

中文关键词：粗糙集;属性约简;RBF网络;算法

外文关键词：Rough Sets ; Attributes Reduction ; RBF Networks ; Algorithms

摘要：粗糙集理论被广泛应用于人工智能、模式识别、数据挖掘和知识发现等领域。而对象的属性约简是是粗糙集理论中的重要问题之一。由于属性约简计算量较大,影响了的粗糙集的实际应用。本文用RBF神经网络高效和OLS对称性的特点,研究粗糙集属性的约简,解决了属性约简的难题,完成了算法的实现,取得了较好的效果。
Rough sets theory was used widely to artificial intelligence, pattern recognition, data mining and knowledge discovery etc fields. Reduction of object attributes is one of important problems of rough sets theory. Actual use of rough sets was affected by more calculations on attribute reduction. In this paper, reduction of rough sets attributes was researched with high efficiency of RBF neural networks and symmetry property of OLS. The difficult problem of attributes reduction was solved by it. Implementation of algorithm was accomplished by it. A better result was gained from it.