CAJ | 학술논문

本文采用人工神经网络的方法对我国第三产业就业比重进行了分析,建立了第三产业就业的BP神经网络模型,以1978～2003年我国人均GDP和第三产业就业比重的历史数据作为样本数据对网络进行训练,通过与回归方程进行比较,发现BP网络在拟合精度上高于回归方程。最后,运用BP网络对不同人均GDP水平下我国第三产业就业比重进行了仿真和测算,指出人工神经网络在经济预测上具有较大的应用价值。
본문채용인공신경망락적방법대아국제삼산업취업비중진행료분석,건립료제삼산업취업적BP신경망락모형,이1978～2003년아국인균GDP화제삼산업취업비중적역사수거작위양본수거대망락진행훈련,통과여회귀방정진행비교,발현BP망락재의합정도상고우회귀방정。최후,운용BP망락대불동인균GDP수평하아국제삼산업취업비중진행료방진화측산,지출인공신경망락재경제예측상구유교대적응용개치。
The paper analyses the proportion of tertiary industry of employment in China using the method of artificial neural network and establishes a BP neural network model of the employment of tertiary industry. Then the network is practiced using GDP per capita from 1978 to 2003 and the proportion of employment in the tertiary industry, and by comparing with regression equation, it is found that the fitting accuracy of the BP network is higher than that of the regression equation. In conclusion, the proportion of employment of the tertiary industry under different GDP per capita level is emulated and estimated using BP network, and the result shows that artificial neural network has great value in economic forecast.