Classification and prediction of whereabouts patterns from the Reality Mining dataset

Classification and prediction of users’ whereabouts patterns is important for many emerging ubiquitous computing applications. Latent Dirichlet Allocation (LDA) is a powerful mechanism to extract recurrent behaviors and high-level patterns (called topics) from mobility data in an unsupervised manner. One drawback of LDA is that it is difficult to give meaningful and usable labels to the extracted topics. We present a methodology to automatically classify the topic with meaningful labels so as to support their use in applications. We also present a topic prediction mechanism to infer user’s future whereabouts on the basis of the extracted topics. Both these two mechanisms are tested and evaluated using the Reality Mining dataset consisting of a large set of continuous data on human behavior.     

Towards a multidimensional view of tourist mobility patterns in cities: A mobile phone data perspective

The last decade has witnessed a wealth of studies on characterizing human mobility patterns using movement datasets. Such efforts have highlighted a few salient dimensions of individual travel behavior relevant to urban planning and policy analysis. Despite the fruitful research outcomes, most of the findings are drawn upon urban residents. The behavioral characteristics of other population groups, such as tourists, remain underexplored. In this study, we introduce an analytical framework to gain insights into tourist mobility patterns. By analyzing mobile phone trajectories of international travelers to three different cities in South Korea, we introduce nine mobility indicators to capture different facets of tourist travel behavior (e.g., duration of stay in a city, spatial extent of activities, location visited and trips conducted, and mobility diversity), and examine their statistical properties across cities. An eigendecomposition approach is then introduced to better understand the interdependency of these mobility indicators and inherent variations among individual travelers. Based on the eigendecomposition results, we further employ a dimension reduction technique to describe the key characteristics of each traveler. Since the mobile phone dataset captures the nationality of tourists, we use such information to quantify the behavioral heterogeneity of travelers across countries and regions. Finally, we select a few traveler groups with distinctive mobility patterns in each city and examine the spatial patterns of their activities. Substantial differences are observed among traveler groups in their spatial preferences. The implications for location recommendation and deployment of tourism services (e.g., transportation) are discussed. We hope the study brings a synergy between classic human mobility analysis and the emerging field of tourism big data. The framework can be applied or extended to compatible datasets to understand travel behavior of tourists, residents, and special population groups in cities.     

Measuring urban regional similarity through mobility signatures

The task of identifying similar regions within and between cities is an important aspect of urban data science as well as applied domains such as real estate, tourism, and urban planning. Regional similarity is typically assessed through comparing socio-demographic variables, resource availability, or urban infrastructure. An essential dimension, often overlooked for this task, is the spatiotemporal mobility patterns of people within a city. In this work we present a novel approach to identifying regional similarity based on human mobility as proxied through micromobility trips. We use a dataset consisting of e-scooter trip origins and destinations for two major European cities that differ in population size and urban structure. Three dimensions of these data are used in modeling the spatial and temporal variability in movement between regions in cities, allowing us to compare regions through a mobility lens. The result is a parameterized similarity model and interactive web platform for comparing regions across different urban environments. The application of this model suggests that human mobility patterns are a quantifiable, unique, and appropriate characteristic through which to measure urban similarity.     

Combining analytics and simulation methods to assess the impact of shared,autonomous electric vehicles on sustainable urban mobility

Urban mobility is currently undergoing three fundamental transformations with the sharing economy, electrification, and autonomous vehicles changing how people and goods move across cities. In this paper, we demonstrate the valuable contribution of decision support systems that combine data-driven analytics and simulation techniques in understanding complex systems such as urban transportation. Using the city of Berlin as a case study, we show that shared, autonomous electric vehicles can substantially reduce resource investments while keeping service levels stable. Our findings inform stakeholders on the trade-off between economic and sustainability-related considerations when fostering the transition to sustainable urban mobility.     

Terrain analysis using radar shape-from-shading

This paper develops a maximum a posteriori (MAP) probability estimation framework for shape-from-shading (SFS) from synthetic aperture radar (SAR) images. The aim is to use this method to reconstruct surface topography from a single radar image of relatively complex terrain. Our MAP framework makes explicit how the recovery of local surface orientation depends on the whereabouts of terrain edge features and the available radar reflectance information. To apply the resulting process to real world radar data, we require probabilistic models for the appearance of terrain features and the relationship between the orientation of surface normals and the radar reflectance. We show that the SAR data can be modeled using a Rayleigh-Bessel distribution and use this distribution to develop a maximum likelihood algorithm for detecting and labeling terrain edge features. Moreover, we show how robust statistics can be used to estimate the characteristic parameters of this distribution. We also develop an empirical model for the SAR reflectance function. Using the reflectance model, we perform Lambertian correction so that a conventional SFS algorithm can be applied to the radar data. The initial surface normal direction is constrained to point in the direction of the nearest ridge or ravine feature. Each surface normal must fall within a conical envelope whose axis is in the direction of the radar illuminant. The extent of the envelope depends on the corrected radar reflectance and the variance of the radar signal statistics. We explore various ways of smoothing the field of surface normals using robust statistics. Finally, we show how to reconstruct the terrain surface from the smoothed field of surface normal vectors. The proposed algorithm is applied to various SAR data sets containing relatively complex terrain structure.     

Practical simulation of virtual crowds using points of interest

We present a computational method that exploits points of interest (POIs) to generate realistic virtual pedestrians for a city model, i.e., a simulated crowd. Our method is validated using mobility traces collected longitudinally from a city-wide free and open Wi-Fi network in downtown Oulu, Finland. Analysing this data, we first construct a time-varying Origin–Destination matrix that describes how individual pedestrians in our city move at different times and places. We compare this ground-truth against a random pedestrian model to investigate how the latter underestimates or overestimates movement at various locations or times of day. By identifying these deviations, we can calibrate a weighted model that uses POIs from OpenStreetMap to adjust the simulated crowd. Our results show a significant accuracy improvement over the random model, while at the same time our work is readily applicable to simulating crowds in other cities (real and virtual) as long as POI can be defined spatially.     

A geographic data science framework for the functional and contextual analysis of human dynamics within global cities

This study develops a Geographic Data Science framework that transforms the Foursquare check-in locations and user origin-destination flows data into knowledge about the emerging forms and characteristics of cities' neighbourhoods. We employ a longitudinal mobility dataset describing human interactions with Foursquare venues in ten global cities: Chicago, Istanbul, Jakarta, London, Los Angeles, New York, Paris, Seoul, Singapore, Tokyo. This social media data provides spatio-temporally referenced digital traces left by human use of urban environments, giving us access to the intangible aspects of urban life, such as people behaviours and preferences. Our framework capitalizes on these new data sources, bringing about a novel Geographic Data Science and human-centered methodological approach. Combining network science – a study area with great promise for the analysis of cities and their structure – with geospatial analysis methods, we model cities as a series of global urban networks. Through a spatially weighted community detection algorithm, we uncover functional neighbourhoods for the ten global cities. Each neighbourhood is linked to hyper-local characterisations of their built environment for the Foursquare venues that compose them, and complemented with a range of measures describing their diversity, morphology and mobility. This information is used in a clustering exercise that uncovers a set of four functional neighbourhood types. Our results enable the profiling and comparison of functional neighbourhoods, based on human dynamics and their contexts, across the sample of global cities. The framework is portable to other geographic contexts where interaction data are available to bind different localities into functional agglomerations, and provide insight into their contextual and human dynamics.     

Land‐use dynamic discovery based on heterogeneous mobility sources

Nowadays, cities are the most relevant type of human settlement and their population has been endlessly growing for decades. At the same time, we are witnessing an explosion of digital data that capture many different aspects and details of city life. This allows detecting human mobility patterns in urban areas with more detail than ever before. In this context, based on the fusion of mobility data from different and heterogeneous sources, such as public transport, transport‐network connectivity and Online Social Networks, this study puts forward a novel approach to uncover the actual land use of a city. Unlike previous solutions, our work avoids a time‐invariant approach and it considers the temporal factor based on the assumption that urban areas are not used by citizens all the time in the same manner. We have tested our solution in two different cities showing high accuracy rates.     

Understanding Skout users’ mobility patterns on a global scale: a data-driven study

Location-based social apps, such as Skout, have been widely used by millions of users for sharing their location information. In this work, we collected all the location information published by over 1.2 million Skout users during December 2012 and June 2016. Based on the collected information, we model the inter-city mobility of Skout users with a global city network, and analyze the evolution of the network based on its structural characteristics. Moreover, we look into Skout users’ mobility patterns by discovering the most popular inter-city routes, destinations, and tightly connected city groups, and analyze the impact on the mobility patterns from geographical distances, languages and cultures. Finally, we leverage machine learning techniques to build a model for identifying the most influential cities in the world according to the Skout data. The results are able to assist individuals, governors and business leaders in making better decisions regarding traveling, immigrating, measuring city improvements and cooperation with cities.

     

Spatio-temporal patterns link your digital identities

An important challenge for mobility analysis is the development of techniques that can associate users’ identities across multiple datasets. These can assist in developing hybrid sensing and tracking mechanisms across large urban spaces, inferring context by combining multiple datasets, but at the same time have important implications for privacy. In this paper we present a scheme to associate different identities of a person across two movement databases. Our two key contributions are the reformulation of this problem in terms of a two-class classification, and the development of efficient techniques for pruning the search space. We evaluate performance of the scheme on synthetic and real data from two co-located city-wide WiFi and Bluetooth networks, and show that the pruning has a remarkable effect on the performance of the scheme in identifying individuals across two distinct mobility datasets. Finally, we discuss the privacy implications of this scheme in the light of our findings.     

On the regularity of human mobility

Understanding human mobility patterns is crucial to fields such as urban mobility and mobile network planning. For this purpose, we make use of large-scale datasets recording individuals spatio-temporal locations, from eight major world cities: Beijing, Tokyo, New York, Paris, San Francisco, London, Moscow and Mexico City. Our contributions are two-fold: first, we show significant similarities in people’s mobility habits regardless of the city and nature of the dataset. Second, we unveil three persistent traits present in an individual’s urban mobility: repetitiveness, preference for shortest-paths, and confinement. These characteristics uncover people’s tendency to revisit few favorite venues using the shortest-path available.     

Spatial context mining approach for transport mode recognition from mobile sensed big data

Knowledge about what transport mode people use is important information of any mobility or travel behaviour research. With ubiquitous presence of smartphones, and its sensing possibilities, new opportunities to infer transport mode from movement data are appearing. In this paper we investigate the role of spatial context of human movements in inferring transport mode from mobile sensed data. For this we use data collected from more than 8000 participants over a period of four months, in combination with freely available geographical information. We develop a support vectors machines-based model to infer five transport modes and achieve success rate of 94%. The developed model is applicable across different mobile sensed data, as it is independent on the integration of additional sensors in the device itself. Furthermore, suggested approach is robust, as it strongly relies on pre-processed data, which makes it applicable for big data implementations in (smart) cities and other data-driven mobility platforms.     

Intelligent bus routing with heterogeneous human mobility patterns

Optimal planning for public transportation is one of the keys helping to bring a sustainable development and a better quality of life in urban areas. Compared to private transportation, public transportation uses road space more efficiently and produces fewer accidents and emissions. However, in many cities people prefer to take private transportation other than public transportation due to the inconvenience of public transportation services. In this paper, we focus on the identification and optimization of flawed region pairs with problematic bus routing to improve utilization efficiency of public transportation services, according to people’s real demand for public transportation. To this end, we first provide an integrated mobility pattern analysis between the location traces of taxicabs and the mobility records in bus transactions. Based on the mobility patterns, we propose a localized transportation mode choice model, with which we can dynamically predict the bus travel demand for different bus routing by taking into account both bus and taxi travel demands. This model is then used for bus routing optimization which aims to convert as many people from private transportation to public transportation as possible given budget constraints on the bus route modification. We also leverage the model to identify region pairs with flawed bus routes, which are effectively optimized using our approach. To validate the effectiveness of the proposed methods, extensive studies are performed on real-world data collected in Beijing which contains 19 million taxi trips and 10 million bus trips.     

Storied spaces: Cultural accounts of mobility, technology, and environmental knowing

When we think of mobility in technical terms, we think of topics such as bandwidth, resource management, location, and wireless networks. When we think of mobility in social or cultural terms, a different set of topics come into view: pilgrimage and religious practice, globalization and economic disparities, migration and cultural identity, daily commutes and the suburbanization of cities.In this paper, we examine the links between these two aspects of mobility. Drawing on non-technological examples of cultural encounters with space, we argue that mobile information technologies do not just operate in space, but they are tools that serve to structure the spaces through which they move. We use recent projects to illustrate how three concerns with mobility and space—legibility, literacy, and legitimacy—open up new avenues for design exploration and analysis.     

A system for destination and future route prediction based on trajectory mining

Location is a key context ingredient and many existing pervasive applications rely on the current locations of their users. However, with the ability to predict the future location and movement behavior of a user, the usability of these applications can be greatly improved. In this paper, we propose an approach to predict both the intended destination and the future route of a person. Rather than predicting the destination and future route separately, we have focused on making prediction in an integrated way by exploiting personal movement data (i.e. trajectories) collected by GPS. Since trajectories contain daily whereabouts information of a person, the proposed approach first detects the significant places where the person may depart from or go to using a clustering-based algorithm called FBM (Forward–Backward Matching), then abstracts the trajectories based on a space partitioning method, and finally extracts movement patterns from the abstracted trajectories using an extended CRPM (Continuous Route Pattern Mining) algorithm. Extracted movement patterns are organized in terms of origin–destination couples. The prediction is made based on a pattern tree built from these movement patterns. With the real personal movement data of 14 participants, we conducted a number of experiments to evaluate the performance of our system. The results show that our approach can achieve approximately 80% and 60% accuracy in destination prediction and 1-step prediction, respectively, and result in an average deviation of approximately 60 m in continuous future route prediction. Finally, based on the proposed approach, we implemented a prototype running on mobile phones, which can extract patterns from a user’s historical movement data and predict the destination and future route.     

一种大数据环境下的在线社交媒体位置推断方法

随着在线社交媒体的快速发展和可定位设备的大量普及,地理位置作为社交媒体大数据中一种质量极高的信息资源,开始在疾病控制、人口流动性分析和广告精准投放等方面得到广泛应用.但是,由于大量用户没有指定或者不能准确指定位置,社交媒体上的地理位置数据十分稀疏.针对此数据稀疏性问题,提出一种基于用户生成内容的位置推断方法UGC-LI(user generate content driven location inference method),实现对社交媒体用户和生成文本位置的推断,为基于位置的个性化信息服务提供数据支撑.通过抽取用户生成文本中的本地词语,构建一个基于词汇地理分布差异和用户社交图谱的概率模型,在多层次的地理范围内推断用户位置.同时,提出一个基于位置的参数化语言模型,计算用户生成文本发出的城市.在真实数据集上进行的评估实验表明:UGC-LI方法能够在15km偏移距离准确定位64.2%的用户,对用户所在城市的推断准确率达到81.3%;同时,可正确定位32.7%的用户生成文本发出的城市,与现有方法相比有明显的提高.     

Energy Optimization under Informed Mobility

Energy optimization is important in wireless ad hoc networks, where node battery power is usually limited. Research results show that such a network can exploit controlled node mobility to reduce communication-related energy consumption. However, node movement itself usually consumes energy. In this paper we study the energy optimization problem that accounts for energy costs associated with both communication and physical node movement. We refer to this model as informed mobility. We first review the theoretical foundations on how to reduce total communication energy consumption, as well as increase system lifetime, by combining node movement and transmission power adaptation. Next, we describe and analyze the informed mobility optimization problem. Based on this analysis, we introduce localized algorithms and protocols for informed mobility. We propose iMobif, a flow-based informed mobility framework that collects network information for mobility decision making. We demonstrate how to use iMobif to minimize total communication energy consumption as well as to maximize system lifetime. We compare the performance of iMobif to that of systems with no mobility or only cost-unaware mobility. Simulation results show iMobif is effective in reducing energy consumption relative to such systems.     

Integrating GPS trajectory and topics from Twitter stream for human mobility estimation

Understanding urban dynamics and large-scale human mobility will play a vital role in building smart cities and sustainable urbanization. Existing research in this domain mainly focuses on a single data source (e.g., GPS data, CDR data, etc.). In this study, we collect big and heterogeneous data and aim to investigate and discover the relationship between spatiotemporal topics found in geo-tagged tweets and GPS traces from smartphones. We employ Latent Dirichlet Allocation-based topicmodeling on geo-tagged tweets to extract and classify the topics. Then the extracted topics from tweets and temporal population distribution from GPS traces are jointly used to model urban dynamics and human crowd flow. The experimental results and validations demonstrate the efficiency of our approach and suggest that the fusion of cross-domain data for urban dynamics modeling is more practical than previously thought.     

Multitask learning and nonlinear optimal control of the COVID-19 outbreak: A geometric programming approach

We propose a multitask learning approach to learn the parameters of a compartmental discrete-time epidemic model from various data sources and use it to design optimal control strategies of human-mobility restrictions that both curb the epidemic and minimize the economic costs associated with implementing non-pharmaceutical interventions. We develop an extension of the SEIR epidemic model that captures the effects of changes in human mobility on the spread of the disease. The parameters of the model are learned using a multitask learning approach that leverages both data on the number of deaths across a set of regions, and cellphone data on individuals’ mobility patterns specific to each region. Using this model, we propose a nonlinear optimal control problem aiming to find the optimal mobility-based intervention strategy that curbs the spread of the epidemic while obeying a budget on the economic cost incurred. We also show that the solution to this nonlinear optimal control problem can be efficiently found, in polynomial time, using tools from geometric programming. Furthermore, in the absence of a straightforward mapping from human mobility data to economic costs, we propose a practical method by which a budget on economic losses incurred may be chosen to eliminate excess deaths due to over-utilization of hospital resources. Our results are demonstrated with numerical simulations using real data from the COVID-19 pandemic in the Philadelphia metropolitan area.     

Modeling and simulating residential mobility in a shrinking city using an agent-based approach

Shrinking cities are characterized by a huge oversupply of dwellings and resulting residential vacancies. Discussions among urban planners and policymakers in Europe have focused on the consequences of urban shrinkage following demographic transition, fertility decline and individualization. In this study, the shrinking city of Leipzig in Eastern Germany is singled out as a case basis for the study of residential mobility and land use change using agent-based modeling techniques, in which social scientists developed a concept of household types based on empirical data that form a unique base; these techniques were used to construct a data-driven, agent-based model. The spatially explicit simulation model RESMOBcity presented here ‘translates’ these empirical data via behavioral rules of households. It computes spatially explicit household patterns, housing demands and residential vacancies. Based on three scenarios, population growth, stagnation and shrinkage, we show that population might stabilize within the coming years. The number of households is expected to further increase. We demonstrate that a selective demolition of vacant housing stock can counteract the enormous oversupply of dwellings and better balance housing demand and the number of available flats. Scenario simulation shows that the model can reproduce observed patterns of population, inner-urban migration and residential vacancy in a spatially explicit manner and thus can be applied to the analysis of scenarios of demographic change in urban regions. The presented model acts as a tool supporting the testing of hypotheses in social science research and allowing the quantification of land-use scenarios in urban regions based on household choices.