基于BA-BP网络构建的企业升级与创新指标参数预测分析

Predictive analysis of enterprise upgrade and innovation indicator parameters based on BA-BP network

针对化工胶粘企业创新创业管理能力预测精度不高,影响企业转型升级的问题,构建了企业创新创业管理能力评价指标体系。采用BA对BP神经网络的初始权值与偏差进行优化,提出了基于BA-BP神经网络的化工胶粘企业创新创业管理能力预测模型。将BA-BP、BP、GA-BP和PSO-BP预测模型分别应用于某省的400家化工胶粘企业中,对比不同预测模型的预测结果。结果表明:BA-BP预测模型对企业创新创业管理能力有更高的预测精度,这为提升化工胶粘企业创新创业能力,实施企业转型升级有参考价值。

Aiming at the problem of low accuracy in predicting the management capability of innovation and entrepreneurship in the adhesive industry,which affects the transformation and upgrading of enterprises,an evaluation index system of enterprise innovation and entrepreneurship management ability was constructed.On this basis,BA is used to optimize the initial weight and deviation of BP neural network and a prediction model of innovation and entrepreneurship management capability of chemical adhesive enterprises based on BA-BP neural network were proposed.BA-BP forecasting model,BP forecasting model,GA-BP forecasting model and4 PSO-BP forecasting model were respectively applied to 400 chemical adhesive enterprises in a province,and the forecasting results of different forecasting models were compared.The results showed that the BA-BP prediction model had higher prediction accuracy for the innovation and entrepreneurship management ability of enterprises,which had certain reference value for improving the innovation and entrepreneurship ability of chemical adhesive enterprises and implementing enterprise transformation and upgrading.