Analysis of human mobility patterns from GPS trajectories and contextual information

Predicting mobility-related behavior is an important yet challenging task. On the one hand, factors such as one’s routine or preferences for a few favorite locations may help in predicting their mobility. On the other hand, several contextual factors, such as variations in individual preferences, weather, traffic, or even a person’s social contacts, can affect mobility patterns and make its modeling significantly more challenging. A fundamental approach to study mobility-related behavior is to assess how predictable such behavior is, deriving theoretical limits on the accuracy that a prediction model can achieve given a specific dataset. This approach focuses on the inherent nature and fundamental patterns of human behavior captured in that dataset, filtering out factors that depend on the specificities of the prediction method adopted. However, the current state-of-the-art method to estimate predictability in human mobility suffers from two major limitations: low interpretability and hardness to incorporate external factors that are known to help mobility prediction (i.e., contextual information). In this article, we revisit this state-of-the-art method, aiming at tackling these limitations. Specifically, we conduct a thorough analysis of how this widely used method works by looking into two different metrics that are easier to understand and, at the same time, capture reasonably well the effects of the original technique. We evaluate these metrics in the context of two different mobility prediction tasks, notably, next cell and next distinct cell prediction, which have different degrees of difficulty. Additionally, we propose alternative strategies to incorporate different types of contextual information into the existing technique. Our evaluation of these strategies offer quantitative measures of the impact of adding context to the predictability estimate, revealing the challenges associated with doing so in practical scenarios.

Download Full-text

Anomalous Urban Mobility Pattern Detection Based on GPS Trajectories and POI Data

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi8070308 ◽

2019 ◽

Vol 8 (7) ◽

pp. 308 ◽

Cited By ~ 3

Author(s):

Zhenzhou Xu ◽

Ge Cui ◽

Ming Zhong ◽

Xin Wang

Keyword(s):

Clustering Algorithm ◽

Human Mobility ◽

Mean Shift ◽

Pattern Detection ◽

Time Interval ◽

Mobility Pattern ◽

Urban Mobility ◽

Mobility Patterns ◽

Functional Regions ◽

Gps Trajectories

Anomalous urban mobility pattern refers to abnormal human mobility flow in a city. Anomalous urban mobility pattern detection is important in the study of urban mobility. In this paper, a framework is proposed to identify anomalous urban mobility patterns based on taxi GPS trajectories and Point of Interest (POI) data. In the framework, functional regions are first generated based on the distribution of POIs by the DBSCAN clustering algorithm. A Weighted Term Frequency-Inverse Document Frequency (WTF-IDF) method is proposed to identify function values in each region. Then, the Origin-Destination (OD) of trips between functional regions is extracted from GPS trajectories to detect anomalous urban mobility patterns. Mobility vectors are established for each time interval based on the OD of trips and are classified into clusters by the mean shift algorithm. Abnormal urban mobility patterns are identified by processing the mobility vectors. A case study in the city of Wuhan, China, is conducted; the experimental results show that the proposed method can effectively identify daily and hourly anomalous urban mobility patterns.

Download Full-text

An Empirical Study of Human Mobility Patterns

10.5753/sbrc.2018.2411 ◽

2018 ◽

Author(s):

Douglas do Couto Teixeira ◽

Jussara M. Almeida

Keyword(s):

Social Network ◽

Empirical Study ◽

Factorial Design ◽

Prediction Models ◽

Human Mobility ◽

Contextual Information ◽

Mobility Prediction ◽

Mobility Patterns ◽

Work Location ◽

Two Phases

This paper documents our efforts towards understanding which factors are more relevant in human mobility prediction. Our work is divided into two phases. First, we characterize a dataset consisting of more than 200,000 user check-ins in the Foursquare social network, inferring important patterns in human mobility. Second, we use factorial design to quantify the importance of several types of contextual information in human mobility prediction. Our results show that the proximity of the users possible next check-in to his or her home and work location are the most important factors (among the ones we analyzed) to be used by mobility prediction models.

Download Full-text

Human mobility data from the BBC Pandemic project

10.1101/2021.02.19.21252079 ◽

2021 ◽

Author(s):

Andrew JK Conlan ◽

Petra Klepac ◽

Adam J Kucharski ◽

Stephen Kissler ◽

Maria L Tang ◽

...

Keyword(s):

Human Mobility ◽

Data Sets ◽

Mobility Patterns ◽

Administrative District ◽

Data Set ◽

Public Science ◽

Mobility Data ◽

The United Kingdom ◽

Television Documentary ◽

Gps Trajectories

AbstractWe present human mobility data for the United Kingdom collected from the “BBC Pandemic”, a public science project linked to the BBC Four television documentary of the same name. Mobile phone GPS trajectories submitted by users and collected over a 24 hour period were aggregated to construct anonymised origin-destination flux matrices at the local administrative district (LAD). We use these data to explore how mobility patterns change with age and employment status - unique stratifications that are not available from other publicly and privately held mobility data sets. We validate the consistency of the aggregated BBC mobility data set against census workflow data and illustrate how the systematic differences in mobility rates with age affect the risk and pattern of transmission between regions with 18-30 year old’s contributing the greatest risk of transmission to adjacent regions, but older 60-100 years playing the most important role in more remote low-density locations.

Download Full-text

Public Transport GPS Probe and Rail Gate Data for Assessing the Pattern of Human Mobility in the Bangkok Metropolitan Region, Thailand

Sustainability ◽

10.3390/su13042178 ◽

2021 ◽

Vol 13 (4) ◽

pp. 2178

Author(s):

Songkorn Siangsuebchart ◽

Sarawut Ninsawat ◽

Apichon Witayangkurn ◽

Surachet Pravinvongvuth

Keyword(s):

Performance Assessment ◽

Public Transport ◽

Human Mobility ◽

Open Data ◽

Capital City ◽

Metropolitan Region ◽

Traffic Information ◽

Mobility Patterns ◽

Transport Planning ◽

Java Program

Bangkok, the capital city of Thailand, is one of the most developed and expansive cities. Due to the ongoing development and expansion of Bangkok, urbanization has continued to expand into adjacent provinces, creating the Bangkok Metropolitan Region (BMR). Continuous monitoring of human mobility in BMR aids in public transport planning and design, and efficient performance assessment. The purpose of this study is to design and develop a process to derive human mobility patterns from the real movement of people who use both fixed-route and non-fixed-route public transport modes, including taxis, vans, and electric rail. Taxi GPS open data were collected by the Intelligent Traffic Information Center Foundation (iTIC) from all GPS-equipped taxis of one operator in BMR. GPS probe data of all operating GPS-equipped vans were collected by the Ministry of Transport’s Department of Land Transport for daily speed and driving behavior monitoring. Finally, the ridership data of all electric rail lines were collected from smartcards by the Automated Fare Collection (AFC). None of the previous works on human mobility extraction from multi-sourced big data have used van data; therefore, it is a challenge to use this data with other sources in the study of human mobility. Each public transport mode has traveling characteristics unique to its passengers and, therefore, specific analytical tools. Firstly, the taxi trip extraction process was developed using Hadoop Hive to process a large quantity of data spanning a one-month period to derive the origin and destination (OD) of each trip. Secondly, for van data, a Java program was used to construct the ODs of van trips. Thirdly, another Java program was used to create the ODs of the electric rail lines. All OD locations of these three modes were aggregated into transportation analysis zones (TAZ). The major taxi trip destinations were found to be international airports and provincial bus terminals. The significant trip destinations of vans were provincial bus terminals in Bangkok, electric rail stations, and the industrial estates in other provinces of BMR. In contrast, electric rail destinations were electric rail line interchange stations, the central business district (CBD), and commercial office areas. Therefore, these significant destinations of taxis and vans should be considered in electric rail planning to reduce the air pollution from gasoline vehicles (taxis and vans). Using the designed procedures, the up-to-date dataset of public transport can be processed to derive a time series of human mobility as an input into continuous and sustainable public transport planning and performance assessment. Based on the results of the study, the procedures can benefit other cities in Thailand and other countries.

Download Full-text

Understanding senior's daily mobility patterns in California using human mobility motifs

Journal of Transport Geography ◽

10.1016/j.jtrangeo.2021.103117 ◽

2021 ◽

Vol 94 ◽

pp. 103117

Author(s):

Rongxiang Su ◽

Jingyi Xiao ◽

Elizabeth C. McBride ◽

Konstadinos G. Goulias

Keyword(s):

Human Mobility ◽

Mobility Patterns ◽

Daily Mobility

Download Full-text

A novel geo-hierarchical population mobility model for spatial spreading of resurgent epidemics

Scientific Reports ◽

10.1038/s41598-021-93810-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Alexandru Topîrceanu ◽

Radu-Emil Precup

Keyword(s):

Computational Models ◽

Human Mobility ◽

Discrete Event ◽

Mobility Model ◽

Hierarchical Organization ◽

Population Mobility ◽

Mobility Patterns ◽

Fine Grained ◽

Epidemic Size ◽

Control Of Epidemics

AbstractComputational models for large, resurgent epidemics are recognized as a crucial tool for predicting the spread of infectious diseases. It is widely agreed, that such models can be augmented with realistic multiscale population models and by incorporating human mobility patterns. Nevertheless, a large proportion of recent studies, aimed at better understanding global epidemics, like influenza, measles, H1N1, SARS, and COVID-19, underestimate the role of heterogeneous mixing in populations, characterized by strong social structures and geography. Motivated by the reduced tractability of studies employing homogeneous mixing, which make conclusions hard to deduce, we propose a new, very fine-grained model incorporating the spatial distribution of population into geographical settlements, with a hierarchical organization down to the level of households (inside which we assume homogeneous mixing). In addition, population is organized heterogeneously outside households, and we model the movement of individuals using travel distance and frequency parameters for inter- and intra-settlement movement. Discrete event simulation, employing an adapted SIR model with relapse, reproduces important qualitative characteristics of real epidemics, like high variation in size and temporal heterogeneity (e.g., waves), that are challenging to reproduce and to quantify with existing measures. Our results pinpoint an important aspect, that epidemic size is more sensitive to the increase in distance of travel, rather that the frequency of travel. Finally, we discuss implications for the control of epidemics by integrating human mobility restrictions, as well as progressive vaccination of individuals.

Download Full-text

Mobility Patterns in Different Age Groups in Japan during the COVID-19 Pandemic: a Small Area Time Series Analysis through March 2021

Journal of Urban Health ◽

10.1007/s11524-021-00566-7 ◽

2021 ◽

Author(s):

Shuhei Nomura ◽

Yuta Tanoue ◽

Daisuke Yoneoka ◽

Stuart Gilmour ◽

Takayuki Kawashima ◽

...

Keyword(s):

Care Delivery ◽

Human Mobility ◽

Age Groups ◽

Virus Transmission ◽

Initial Response ◽

Health Care Delivery System ◽

Population Characteristics ◽

Mobility Patterns ◽

Mobility Data ◽

The Status

AbstractIn the COVID-19 era, movement restrictions are crucial to slow virus transmission and have been implemented in most parts of the world, including Japan. To find new insights on human mobility and movement restrictions encouraged (but not forced) by the emergency declaration in Japan, we analyzed mobility data at 35 major stations and downtown areas in Japan—each defined as an area overlaid by several 125-meter grids—from September 1, 2019 to March 19, 2021. Data on the total number of unique individuals per hour passing through each area were obtained from Yahoo Japan Corporation (i.e., more than 13,500 data points for each area). We examined the temporal trend in the ratio of the rolling seven-day daily average of the total population to a baseline on January 16, 2020, by ten-year age groups in five time frames. We demonstrated that the degree and trend of mobility decline after the declaration of a state of emergency varies across age groups and even at the subregional level. We demonstrated that monitoring dynamic geographic and temporal mobility information stratified by detailed population characteristics can help guide not only exit strategies from an ongoing emergency declaration, but also initial response strategies before the next possible resurgence. Combining such detailed data with data on vaccination coverage and COVID-19 incidence (including the status of the health care delivery system) can help governments and local authorities develop community-specific mobility restriction policies. This could include strengthening incentives to stay home and raising awareness of cognitive errors that weaken people's resolve to refrain from nonessential movement.

Download Full-text