作者: Brian Don Tate
DOI:
关键词: Neural gas 、 Node (networking) 、 Artificial neural network 、 Network simulation 、 Data mining 、 Probabilistic neural network 、 Brooks–Iyengar algorithm 、 Time delay neural network 、 Algorithm 、 Relational database 、 Computer science
摘要: A method, apparatus, and article of manufacture for performing data mining applications in a relational database management system. At least one analytic algorithm enhanced back-propagation neural network processing is performed by computer, wherein the includes SQL statements system directly against programmatic iteration. The operates on that has been partitioned into training, testing validation sets. maps training sets to nodes processed as it moves from an input node through hidden output network. In addition, determines error difference between node's value target mapped network, changes weight or more based accumulation node, order get converge solution. Finally, cross-validates changed prevent overfitting node.