基于改进线性回归模型的民航货物周转量预测

Prediction of civil aviation cargo turnover based on improved linear regression model

采用遗传算法对线性回归模型中的求参过程进行优化,以期提高民航货物周转量预测模型精度。首先运用线性回归模型对民航货物周转量进行预测,结果预测准确率为0. 83;之后采用遗传算法进行优化,且当交叉默认概率Pc=0. 85,变异概率为Pm=0. 01,所得预测结果最优,准确率为0. 94。结果表明:利用遗传算法对线性回归模型进行优化能够提高民航货物周转量预测精度。

In this paper,genetic algorithm was used to optimize the parameter extraction process in the linear regression model in order to improve the accuracy of the prediction model of civil aviation cargo turnover. Firstly,the linear regression model was used to predict the cargo turnover of civil aviation,and the prediction accuracy was 0. 83. After that,genetic algorithm was adopted for optimization,and when the default crossover probability Pc= 0. 85 and the mutation probability Pm= 0. 01,the predicted result was optimal with accuracy of 0. 94. The results show that the optimization of linear regression model by genetic algorithm can improve the prediction accuracy of civil aviation cargo turnover.