基于深度置信网络的车辆交通流预测-《集美大学学报（自然科学版）》

文章信息/Info

作者:: 付永钢1; 李传目1; 王惠蓉2; (1.集美大学计算机工程学院，福建厦门 361021;2.集美大学海洋文化与法律学院，福建厦门 361021)

Author(s):: FU Yonggang1; LI Chuanmu1; WANG Huirong2; (1.College of Computer Engineering,Jimei University,Xiamen 361021,China;2.College of Marine Culture and Law,Jimei University,Xiamen 361021,China)

Keywords:: deep belief networks; traffic flow prediction; machine learning; deep learning

摘要:: 提出了一种新的基于深度置信网络的交通流预测方法，利用深度置信网络良好的训练和预测性能，能够很好地学习时序数据集的内部特征，从而准确地预测交通数据流。为了验证算法的有效性，在PeMS数据集上对算法进行了实验测试，并同其他相关预测和分析方法进行了比较，实验结果表明新算法具有较好的预测性能。

Abstract:: In this paper,a novel traffic flow prediction method based on deep belief network(DBN) is investigated and compared with related methods.Due to the good learning ability of DBN network,the proposed DBN based traffic flow prediction method can learn the internal features well,and predict traffic data flow more accurately.The proposed method is tested on the PeMS dataset,and the experimental results show that the proposed method has good prediction performance.