广义回归神经网络在全国高技能人才需求中的应用研究

Research on Application of Generalized Regression Neural Network in the National Demand of High-skilled Talents

摘要:

广义回归神经网络（GRNN）克服了BP神经网络运行中收敛速度慢且局部极小的不足，正逐渐被应用到货运量、农业需求及煤炭需求方面，文章引入GRNN神经网络对全国高技能人才需求进行预测检验，并从五个角度将其与GM（1，1）模型的输出结果做对比分析,结果表明GRNN模型具有优异的预测性能，验证了把该模型引入高技能人才需求这一新领域是有效可行的，为今后进一步定量研究有关高技能人才需求方面提供一种新方法。

Abstract:

Generalized Regression Neural Network (GRNN) overcomes the defects of BP neural network which are slow convergence speed and local minimum and so forth. It is gradually being applied to freight volume, agricultural demand and coal demand. This paper attempts to introduce GRNN to forecast and test the demand of national high-skilled talents.Then from five angles,it analyses and compares the outputs of GRNN with those of GM(1,1) model.The result shows that GRNN model has superior prediction performance and proves that introducing the model into this new field about demand for high-skilled talents is effective and feasible,which provides a new method for the further quantitative research on demand of high-skilled talents in the future.