Forecasting urban dynamics with mobility logs by bilinear Poisson regression

Abstract

Understanding people flow in a city (urban dynamics) is of great importance in urban planning, emergency management, and commercial activity. With the spread of smart devices, many studies on urban dynamics modeling with mobility logs have been conducted. It is predictive analysis, not analysis of the past, that enables various applications contributing to a more prosperous society. To deal with the non-linear effects on urban dynamics from external factors, such as day of the week, national holiday, or weather, we propose a low-rank bilinear Poisson regression model, for a novel and flexible representation of urban dynamics predictive analysis. The results obtained from an experiment with one year's worth of mobility records suggest the high prediction accuracy of the proposed model. We also introduce the following applications: regional event detection via irregularities, visualization of urban dynamics corresponding to urban demographics, and extraction of urban demographics of unknown point of interests.

Detecting interpersonal relationships in large-scale railway trip data

“…The method proposed in the present study, which detects friends without accidentally including people in the same activity group, is applicable to other datasets if these data were recorded over a sufficiently long period. In previous studies, mobility was regulated and restricted by the relationships between places and humans [9,15,22,24,27]. On a broader scale, human mobility can be understood as an aspect of human-human relationships and place-human relationships, because many trips are conducted for the sole purpose of meeting friends.…”

“…Understanding and predicting human mobility is important for many applications such as preventing the spread of disease [1], transportation engineering [13,28], and event planning [24]. Most previous models focused on human mobility behavior [10,27] and interactions between humans and the characteristics of a place [9,12,15,22].…”

“…By contrast, human trajectories collected over a large area can better represent daily human activities. For people living in large cities, trip data from railway systems reflect their activities of daily life [24]. Connecting interpersonal relationship data with human relationships helps us better understand daily human interactions.…”

Predicting Transportation Carbon Emission with Urban Big Data

“…Fan [9] proposed a predicting-by-clustering framework to predict crowd behavior at a citywide level based on human mobility big data. Shimosaka [10] proposed a low-rank bilinear Poisson regression model to predict the urban dynamics, utilizing the one year's worth of mobility records. Li [11] proposed a smart city infrastructure to serve people, based on the multisensors data.…”

“…As described in the Table 1, we divide the POI dataset into twelve classes, i.e., {C 1 , C 2 , · · · , C 12 }. The density of POIs for the classification C i in the affecting region R of the grid g can be formulated by Equation 10.…”