首页>
外国专利>
SEARCH CLUSTER OPTIMISATION METHOD AND SYSTEM BASED ON RBF NEURAL NETWORK
SEARCH CLUSTER OPTIMISATION METHOD AND SYSTEM BASED ON RBF NEURAL NETWORK
展开▼
机译:基于RBF神经网络的搜索集群优化方法及系统
展开▼
页面导航
摘要
著录项
相似文献
摘要
A search cluster optimisation method and system based on an RBF neural network, relating to the field of big data searches. The system comprises a data node, a management node and an RBF neural network training unit. The RBF neural network training unit carries out a cluster search based on an RBF neural network, regularly performs training and calculation by taking data, such as the number of indexes of the data node, search feedback, CPU search consumption, a search response duration and a node load pressure, as an input vector set, sets a self-defined error range, the number of iterations and a weight value, and performs training to obtain a cluster and a management node most suitable therefor. When a certain management node fails, a system automatically takes out training result data of a failed data node under a cluster of the management node, and finds a management node suitable for the failed node, establishes a connection request procedure between the failed data node and the management node, and automatically applies to join in the cluster.
展开▼