摘要
随着数据挖掘技术的兴起,为了提高数据挖掘的准确性,提出了很多数据挖掘算法。神经网络与粗糙集理论结合的数据挖掘算法一直是基于粗糙集理论数据挖掘研究的热点之一。文中提出利用RBF神经网络收敛速度快、泛化能力强等优势先对数据进行训练,优化数据后传递给粗糙集进行数据挖掘的新思路。并通过对比与未经过RBF神经网络训练的数据挖掘结果,发现RBF神经网络与粗糙集结合算法挖掘的精度有明显的提高,证明了RBF神经网络与粗糙集理论结合的数据挖掘算法是有效的、可行的。
With the rise of data mining technology,in order to improve the accuracy of data mining , a lot of data mining algorithms have been put forward. The data mining algorithm which combinesneural networks with rough set theory has been one of the hot spots of data mining research based on rough set theory. Put forward the new idea of training data firstly, then pass to rough sets data mining after elim- inating interference data, take the advantages of Radical Basis Function (RBF) neural network:fast convergence rate and strong generali- zation capability etc. And through the contrast to the data mining results which not using RBF neural network training, the precision of the algorithm which combined RBF neural network with rough set is greatly improved,it shows that the data mining algorithm which com- bined with neural network and rough sets theory has validity and feasibility.
出处
《计算机技术与发展》
2013年第7期87-91,共5页
Computer Technology and Development
基金
国家自然科学基金资助项目(61100116/F020512)