FMA-ETA: Estimating Travel Time Entirely Based on FFN with Attention
FMA-ETA: Estimating Travel Time Entirely Based on FFN with Attention
复制标题
DOI：
10.1109/icassp39728.2021.9414054
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
Yiwen Sun;Yulu Wang;Kun Fu;Zheng Wang;Ziang Yan;Changshui Zhang;Jieping Ye
中科院分区：
文献类型：
--
作者：
Yiwen Sun;Yulu Wang;Kun Fu;Zheng Wang;Ziang Yan;Changshui Zhang;Jieping Ye
估计到达时间(ETA)是智能交通系统(ITS)中最重要的服务之一，是近年来具有挑战性的时空数据挖掘任务。目前，基于深度学习的方法，特别是基于递归神经网络(RNN)的方法，被用于从海量数据中对ETA的ST段模式进行建模，成为最新的研究方向。然而，RNN的结构不利于并行计算，存在训练和推理速度慢的问题。为了解决这一问题，我们提出了一种新颖、简洁和有效的框架，主要基于前馈网络(FFN)的ETA和具有多因素注意的FFN(FMA-ETA)。提出了一种新的多因素注意机制，用于处理不同的类别特征，并有目的地聚合信息。在真实车辆行驶数据集上的大量实验结果表明，FMA-ETA在预测精度和推理速度方面与最先进的方法相媲美。
Estimated time of arrival (ETA) is one of the most important services in intelligent transportation systems (ITS) and becomes a challenging spatial-temporal (ST) data mining task in recent years. Nowadays, deep learning based methods, specifically recurrent neural networks (RNN) based ones are adapted to model the ST patterns from massive data for ETA and become the state-of-the-art. However, RNN is suffering from slow training and inference speed, as its structure is unfriendly to parallel computing. To solve this problem, we propose a novel, brief and effective framework mainly based on feed-forward network (FFN) for ETA, FFN with Multifactor Attention (FMA-ETA). The novel Multi-factor Attention mechanism is proposed to deal with different category features and aggregate the information purposefully. Extensive experimental results on the real-world vehicle travel dataset show FMA-ETA is competitive with state-of-the-art methods in terms of the prediction accuracy with significantly better inference speed.