The Performance Evaluation Of Industrial Cluster Based On Neural Network

As an industrial economic development stage of certain senior phase, the release of industry cluster performance directly affects the regional economy comprehensive strength in the medium level. Using neural network method to evaluate performance of cluster,research of industry cluster external economy effect, will enrich the performance of industrial cluster analysis method, and help to define the development of regional industry cluster phase, provides the technical support for industrial policy.First of all, comprehensive analyzes the related data of industrial cluster at home and abroad, defines the connotation of performance of industrial cluster category, analyzes the internal mechanism of industrial cluster performance, has paved the way for the cluster performance comprehensive evaluation.In the next place, combining with the characteristics of high and new technology industry.From the cluster size, cluster efficiency, technological innovation ability, the support structure and project construction five dimensions, designs the clustering performance evaluation index system.After the dimensionless processing to index data, its processing result input to neural network.Once more, using the comprehensive evaluation model of BP network method and self-organizing competitive network as a case analysis. By principal components analysis to obtain the target of BP neural networks, using the sample data of 25 provinces to finish BP network training and learning, using simulation to achieve precision of BP network.The industry cluster performance indicators data of Beijing,Tianjin and Hebei will be the input source, samples for evaluation of the performance evaluation of industrial cluster results; From the sample data by the method of self-organizing competitive network classify performance evaluation, get the high-tech industrial cluster performance level classification results of Beijing,Tianjin and Hebei.Finally, the comprehensive analysis of the organizational learning classification results and the evaluation value of BP network, determine the region cluster performancestage. The result analysis shows that Beijing's high-tech industrial cluster performance between medium and medium to high; Hebei and Tianjin degree of performance evaluation results are the lowest. The evaluation value of the BP network was also a range of the class, so the effective comprehensive evaluation results.In this paper, the integrated use of BP neural network and the self-organizing competitive network method, analyzes the quantitative evaluation value of BP network and the classification of self-organizing competitive network defined, implements the effective analysis of the performance of industrial cluster degree.From the perspective of performance show industry development policy,and put forward related suggestions.