Unsupervised Learning for Human Mobility Behaviors

2021 ◽  
Author(s):  
Siyuan Liu ◽  
Shaojie Tang ◽  
Jiangchuan Zheng ◽  
Lionel M. Ni
Keyword(s):  
Unsupervised Learning ◽  
Human Mobility ◽  
Data Sets ◽  
Individual Data ◽  
Context Reasoning ◽  
Mobility Data ◽  
Temporal And Spatial ◽  
Mobile Context ◽  
Location Sensing ◽  
Spatial Sparsity

Learning human mobility behaviors from location-sensing data are crucial to mobility data mining because of its potential to address a range of analytical purposes in mobile context reasoning, including exploration, inference, and prediction. However, existing approaches suffer from two practical problems: temporal and spatial sparsity. To address these shortcomings, we present two unsupervised learning methods to model the mobility behaviors of multiple users (i.e., a population), considering efficiency and accuracy. These methods intelligently overcome the sparsity in individual data by seeking temporal commonality among users’ heterogeneous location behaviors. The advantages of our models are highlighted through experiments on several real-world mobility data sets, which also show how our methods can realize the three analytical purposes in a unified manner.

scholarly journals Human mobility data from the BBC Pandemic project

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.

scholarly journals EXPLORING SIMILARITIES AND VARIATIONS OF HUMAN MOBILITY PATTERNS IN THE CITY OF LONDON

2018 ◽  
Vol XLII-4/W11 ◽  
pp. 51-58 ◽  
Author(s):  
P. Sulis ◽  
E. Manley
Keyword(s):  
Spatial Data ◽  
Human Mobility ◽  
Temporal Trends ◽  
Temporal Variations ◽  
Quantitative Information ◽  
Data Sets ◽  
Mobility Patterns ◽  
Mobility Data ◽  
Spatial Differences ◽  
Using Data

<p><strong>Abstract.</strong> The availability of new spatial data represents an unprecedented opportunity to better understand and plan cities. In particular, extensive data sets of human mobility data supply new information that can empower urbanism research to unveil how people use and visit urban places over time, overcoming traditional limitations related to the lack of large, detailed data sets. In this work, we explore patterns of similarities and spatial differences in human mobility flows in London, analysing their temporal variations in relation to the liveliness measured in a number of places. Using data sourced from the Oyster smart card and Twitter, we perform a time-series cluster analysis, exploring the similarity of temporal trends amongst places assigned to each cluster. Results suggest that differences in patterns appear to be related to the central and peripheral location of places, which present two or more temporal trends over the week. The type of transport network connecting the places (Tube, Railways, etc.) also appears to be a factor in determining significant differences. This work contributes to current urbanism research investigating the daily rhythms in cities. It also explores how to use mobility data to classify places according to their temporal features, with the aim of enhancing conventional analysis tools and integrating them with new quantitative information and methods.</p>

scholarly journals Detecting and Reducing Biases in Cellular-Based Mobility Data Sets

Entropy ◽  
2018 ◽  
Vol 20 (10) ◽  
pp. 736
Author(s):  
Alicia Rodriguez-Carrion ◽  
Carlos Garcia-Rubio ◽  
Celeste Campo
Keyword(s):  
Mobile Phone ◽  
Mobile Networks ◽  
Human Mobility ◽  
Mobility Model ◽  
Data Sets ◽  
Location Data ◽  
Mobility Data ◽  
Key Factor ◽  
Data Collection Approach ◽  
Call Detail Records

Correctly estimating the features characterizing human mobility from mobile phone traces is a key factor to improve the performance of mobile networks, as well as for mobility model design and urban planning. Most related works found their conclusions on location data based on the cells where each user sends or receives calls or messages, data known as Call Detail Records (CDRs). In this work, we test if such data sets provide enough detail on users’ movements so as to accurately estimate some of the most studied mobility features. We perform the analysis using two different data sets, comparing CDRs with respect to an alternative data collection approach. Furthermore, we propose three filtering techniques to reduce the biases detected in the fraction of visits per cell, entropy and entropy rate distributions, and predictability. The analysis highlights the need for contextualizing mobility results with respect to the data used, since the conclusions are biased by the mobile phone traces collection approach.

scholarly journals Activity Space Maps: a novel human mobility data set for quantifying time spent at risk

2021 ◽  
Author(s):  
Daniel T Citron ◽  
Shankar Iyer ◽  
Robert C Reiner ◽  
David L Smith
Keyword(s):  
Human Mobility ◽  
Epidemiological Data ◽  
Global Scale ◽  
Activity Space ◽  
Data Sets ◽  
Data Set ◽  
Epidemiological Modeling ◽  
Vector Borne Diseases ◽  
Geographic Space ◽  
Mobility Data

Activity Space Maps are a novel global-scale movement and mobility data set which describes how people distribute their time through geographic space. The maps are intended for use by researchers for the purposes of epidemiological modeling. Activity Space Maps are designed to complement existing digitally-collected mobility data sets by quantifying the amount of time that people spend in different locations. This information is important for estimating the duration of contact with the environment and the potential risk of exposure to disease. More concretely, the type of information contained in Activity Space Maps will make it easier to model the spatial transmission patterns of vector-borne diseases like malaria and Dengue fever. We will discuss the motivation for designing Activity Space Maps, how the maps are generated from mobile phone user app location history data, and discuss an example use case demonstrating how such data may be used together with spatial epidemiological data to advance our understanding of spatial disease patterns and the relationship between travel behaviors and infection risk.

scholarly journals Observing the silent world under COVID-19 with a comprehensive impact analysis based on human mobility

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Shaobin Wang ◽  
Yun Tong ◽  
Yupeng Fan ◽  
Haimeng Liu ◽  
Jun Wu ◽  
...  
Keyword(s):  
Impact Analysis ◽  
Human Mobility ◽  
Economic Activities ◽  
Underdeveloped Countries ◽  
Mobility Data ◽  
Long Run ◽  
Level Of Economic Development ◽  
Human World ◽  
Dynamic Relationships ◽  
The Impact

AbstractSince spring 2020, the human world seems to be exceptionally silent due to mobility reduction caused by the COVID-19 pandemic. To better measure the real-time decline of human mobility and changes in socio-economic activities in a timely manner, we constructed a silent index (SI) based on Google’s mobility data. We systematically investigated the relations between SI, new COVID-19 cases, government policy, and the level of economic development. Results showed a drastic impact of the COVID-19 pandemic on increasing SI. The impact of COVID-19 on human mobility varied significantly by country and place. Bi-directional dynamic relationships between SI and the new COVID-19 cases were detected, with a lagging period of one to two weeks. The travel restriction and social policies could immediately affect SI in one week; however, could not effectively sustain in the long run. SI may reflect the disturbing impact of disasters or catastrophic events on the activities related to the global or national economy. Underdeveloped countries are more affected by the COVID-19 pandemic.

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

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.

scholarly journals Comparative Analysis of Geolocation Information through Mobile-Devices under Different COVID-19 Mobility Restriction Patterns in Spain

2021 ◽  
Vol 10 (2) ◽  
pp. 73
Author(s):  
Raquel Pérez-Arnal ◽  
David Conesa ◽  
Sergio Alvarez-Napagao ◽  
Toyotaro Suzumura ◽  
Martí Català ◽  
...  
Keyword(s):  
Mobile Phones ◽  
Human Mobility ◽  
The Other ◽  
Data Sources ◽  
Experimental Setting ◽  
Mobile Data ◽  
Restriction Patterns ◽  
Mobility Data ◽  
The World ◽  
Proper Interpretation

The COVID-19 pandemic is changing the world in unprecedented and unpredictable ways. Human mobility, being the greatest facilitator for the spread of the virus, is at the epicenter of this change. In order to study mobility under COVID-19, to evaluate the efficiency of mobility restriction policies, and to facilitate a better response to future crisis, we need to understand all possible mobility data sources at our disposal. Our work studies private mobility sources, gathered from mobile-phones and released by large technological companies. These data are of special interest because, unlike most public sources, it is focused on individuals rather than on transportation means. Furthermore, the sample of society they cover is large and representative. On the other hand, these data are not directly accessible for anonymity reasons. Thus, properly interpreting its patterns demands caution. Aware of that, we explore the behavior and inter-relations of private sources of mobility data in the context of Spain. This country represents a good experimental setting due to both its large and fast pandemic peak and its implementation of a sustained, generalized lockdown. Our work illustrates how a direct and naive comparison between sources can be misleading, as certain days (e.g., Sundays) exhibit a directly adverse behavior. After understanding their particularities, we find them to be partially correlated and, what is more important, complementary under a proper interpretation. Finally, we confirm that mobile-data can be used to evaluate the efficiency of implemented policies, detect changes in mobility trends, and provide insights into what new normality means in Spain.

Sign in / Sign up

Export Citation Format

Share Document