期刊名称：International Journal of Computer Trends and Technology
电子版ISSN：2231-2803
出版年度：2013
卷号：4
期号：9-3
出版社：Seventh Sense Research Group
摘要：A common problem in bio medical data using neural networks for classification purposes are complex nature of the data, high dimensionality, convoluted and overlapping classes with their remarkable ability to derive meaning from complicated data, can be extract patterns and trends are too complex. Bio medical classification is a complex process to make decisions. Classification performance of the neural network suffers dramatically. This paper proposes neural network and data mining techniques are combined to automate biomedical classification processes to support decision. To improve the classification ability and behavior of neural network is used by preprocessing and preclustered data with the help of Rule based induction, Multilayer perceptron model, nearest neighbor, Radial basics function and back propagation learning algorithm is employed to classify such complex tasks. The proposed clustering algorithm applied to the bio medical dataset to reduce the amount of samples to be presented to the neural network. It improves accuracy and computation time when applied to the publicly available benchmark bio medical dataset.
关键词：Neural Network; Data mining; back propagation; Multilayer perceptron
