基于随机森林回归的航班延误预测
|
摘要
随着国内民航领域的飞速发展,航班延误次数及程度也在不断加深。因此,各航空公司对于航班延误时间预测的需求就更加强烈。采用随机森林回归算法来进行航班延误的预测。其中使用的原始数据来自于美国交通统计局(BST)上发布的数据。首先,对原始数据进行处理,通过分析影响航班到港时间的因素进行特征筛选,并进行数据清洗。然后,训练模型,并使用Grid-Search和交叉验证法选取最优的参数。最后,与支持向量机回归与岭回归算法进行对照分析,发现随机森林有较好的预测效果。实验结果显示,延误预测的R-squared为0.91和平均绝对误差为10.56分钟。
With the rapid development of the civil aviation industry, the phenomenon of flight delays has become more frequent. Therefore, forecasting flight delays is particularly important. Uses the random forest regression algorithm to predict flight delays. The raw data used in the search is derived from data published by The Bureau of Transportation Statistics(BST). First, processes the raw data, and analyzes the characteristics of the flight arrival time to analyze the characteristics and do the data cleaning. Then, trains the model and uses the Grid-Search and cross-validation to select the optimal parameters. Finally, compared with support vector machine regression and ridge regression algorithm,it is found that random forest has better prediction results. The experimental results show that R2 is 0.91 and MAE is 10.56 minutes.
With the rapid development of the civil aviation industry, the phenomenon of flight delays has become more frequent. Therefore, forecasting flight delays is particularly important. Uses the random forest regression algorithm to predict flight delays. The raw data used in the search is derived from data published by The Bureau of Transportation Statistics(BST). First, processes the raw data, and analyzes the characteristics of the flight arrival time to analyze the characteristics and do the data cleaning. Then, trains the model and uses the Grid-Search and cross-validation to select the optimal parameters. Finally, compared with support vector machine regression and ridge regression algorithm,it is found that random forest has better prediction results. The experimental results show that R2 is 0.91 and MAE is 10.56 minutes.
引文
[1]曹卫东,林翔宇.基于贝叶斯网络的航班过站时间分析与延误预测[J].计算机工程与设计,2011,32(5):1170-1172.
[2]Rebollo J J,Balakrishnan H. Characterization and Prediction of Air Traffic Delays[J]. Transportation Research Part C:Emerging Technologies,2014.
[3]罗赟骞,陈志杰,汤锦辉,等.采用支持向量机回归的航班延误预测研究[J].交通运输系统工程与信息,2015,15(1):143-149.
[4]罗凤娥,张成伟.基于时序数据挖掘的航班延误预测分析[J].现代电子技术,2014,37(24):52-55.
[5]程华,李艳梅,罗谦,等.基于C4.5决策树方法的到港航班延误预测问题研究[J].系统工程理论与实践,2014,34(增1):239-247.
[2]Rebollo J J,Balakrishnan H. Characterization and Prediction of Air Traffic Delays[J]. Transportation Research Part C:Emerging Technologies,2014.
[3]罗赟骞,陈志杰,汤锦辉,等.采用支持向量机回归的航班延误预测研究[J].交通运输系统工程与信息,2015,15(1):143-149.
[4]罗凤娥,张成伟.基于时序数据挖掘的航班延误预测分析[J].现代电子技术,2014,37(24):52-55.
[5]程华,李艳梅,罗谦,等.基于C4.5决策树方法的到港航班延误预测问题研究[J].系统工程理论与实践,2014,34(增1):239-247.