阶层分化失地农民风险研究的新视角

《阶层分化失地农民风险研究的新视角》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、0.770.120.520.520.730.520.090.150.340.240.390.420.710.450.240.690.320.220.120.650.750.010.660.050.000.810.680.440.520.690.490.120.330.360.100.520.640.480.090.330.600.390.610.570.680.290.660.210.260.480.580.350.550.650.200.130.560.

2、390.020.450.410.530.330.920.030.080.370.440.440.480.430.370.370.330.480.470.800.550.320.120.040.760.830.760.130.490.650.510.560.220.30Frontier前沿汇0.010.800.820.520.060.020.540.020.540.330.040.160.85-11.721.571.150.920.640.540.29

3、0.020.240.680.701.171.351.341.720.290.940.450.520.800.410.010.130.150.260.690.790.250.600.08为13-15-1型三层BP神经网络专家系统。表3神经网络测试数据采用MATLAB神经网络工具箱对网络进行训练，隐层训练函数为tansig，输出层训练函数为purelin；为保证训练结果的收敛，神经网络训人员学国外人才国家级省部级院所级SCIEI核心实用发明国家级省部级预测值编号历经历培养项目项目项目论文论文论文新型专利奖项奖项练次

4、数设置为1000次；为提高训练的精度，训练误差设置为1E-7，其余参数按默认设置。在确定了网络的结构后，要通过输入和输出样本集对网络进行训练，11309016255436559827730172.88150即对网络的阈值和权值进行学习和修正，以使网络实现既定的输入输出映射关系。当训练误差达到了精度要求，训练停止。训练后的神经网络模型结构如下。114090251061913201061041.71790输入层到隐层的权值矩阵为：0.420.740.300.700.010.710.410.030.560.160.520.28

5、0.5600.8.43200.2.77410.0.30000..70010.01.7010.01.57100.1.43100.0.0330.00.4560.707.160.120.520.450.280.520.5600..823300..724710..050700.2.05100.4.1970.707.150.508.130.408.030.302.040.207.770.206.120.670.450.100.52150008554152036040101.0744

6、00.2.43352000..77.714400.3.4506700.74.2205000.4..409100.0.37.1771000.1..54281000.4..40830.2030.5.3620.7500..21760.8020.2.5620.5100.6.2780.5300.1.5060.982300.2.77301..60400.206.0100.1.1670.0.8150.503.130.302.030.150.040.608.770.209.120.4

7、90.450.040.520.350.710.460.420.400.310.120.400.230.750.820.510.5300..962723000.7..2743300.0.06.154700.01.2.326500..54129600.0.17.70780.008..15583000.3..342280.010.61.3520.7900..26780.5001..22690.4200..64790.6000..1004可行且有效的。00..534500

8、..74160.07.2460.05.54200.0.4900.606.310.103.120.102.400.08.0230.02.5750.109.820.140.510.150.5