Predicting Urban Dispersal Events: A Two-Stage Framework through Deep Survival Analysis on Mobility Data
Predicting Urban Dispersal Events: A Two-Stage Framework through Deep Survival Analysis on Mobility Data
复制标题
DOI：
10.1609/aaai.v33i01.33015199
复制
发表时间：
2019-05
期刊：
影响因子：
--
通讯作者：
Amin Vahedian;Xun Zhou;Ling Tong;W. Street;Yanhua Li
中科院分区：
文献类型：
--
作者：
Amin Vahedian;Xun Zhou;Ling Tong;W. Street;Yanhua Li
城市疏散事件是指在短时间内大量人口离开同一地区的过程。疏散事件的早期预测对于缓解拥堵和安全风险以及为出租车和共乘车队做出更好的调度决策非常重要。现有的工作主要集中在预测出租车需求在不久的将来从历史数据的学习模式。然而，它们在异常情况下失败，因为具有异常高需求的分散事件是非重复的，并且违反了常见的假设，例如需求随时间变化的平滑性。相反，在本文中，我们认为，分散事件遵循一个复杂的模式，在过去的行程和其他相关功能，可以用来预测这样的事件。因此，我们制定的扩散事件预测问题的生存分析问题。我们提出了一个两阶段框架（DILSA），其中开发了一个结合生存分析的深度学习模型来预测扩散事件的概率及其需求量。我们对2014年至2016年的纽约市黄色出租车数据集进行了广泛的案例研究和实验。结果表明，DILSA可以预测未来5小时内的事件，F1评分为0：7，平均时间误差为18分钟。它比最先进的出租车需求预测深度学习方法好几个数量级。
Urban dispersal events are processes where an unusually large number of people leave the same area in a short period. Early prediction of dispersal events is important in mitigating congestion and safety risks and making better dispatching decisions for taxi and ride-sharing fleets. Existing work mostly focuses on predicting taxi demand in the near future by learning patterns from historical data. However, they fail in case of abnormality because dispersal events with abnormally high demand are non-repetitive and violate common assumptions such as smoothness in demand change over time. Instead, in this paper we argue that dispersal events follow a complex pattern of trips and other related features in the past, which can be used to predict such events. Therefore, we formulate the dispersal event prediction problem as a survival analysis problem. We propose a two-stage framework (DILSA), where a deep learning model combined with survival analysis is developed to predict the probability of a dispersal event and its demand volume. We conduct extensive case studies and experiments on the NYC Yellow taxi dataset from 20142016. Results show that DILSA can predict events in the next 5 hours with F1-score of 0:7 and with average time error of 18 minutes. It is orders of magnitude better than the state-of-the-art deep learning approaches for taxi demand prediction.