Real-world Mobile Data Research Articles

Distributed versus centralized computing of coverage in mobile crowdsensing

The expected spatial coverage of a crowdsensing platform is an important parameter that derives from the mobility data of the crowdsensing platform users. We tackle the challenge of estimating the anticipated coverage while adhering to privacy constraints, where the platform is restricted from accessing detailed mobility data of individual users. Specifically, we model the coverage as the probability that a user detours to a point of interest if the user is present in a certain region around that point. Following this approach, we propose and evaluate a centralized as well as a distributed implementation model. We examine real-world mobility data employed for assessing the coverage performance of the two models, and we show that the two implementation models provide different privacy requirements but are equivalent in terms of their outputs.

Like it, buy it? Examining the role of bookmarking in the mediation of visual appeal and purchase intent from a dual-system perspective

PurposeWhile prior studies have explored the relationship between visual appeal and purchasing decisions, the role of bookmarking has largely been underemphasized. This research aims to address this gap by focusing on the impact of bookmarking on consumer behavior, guided by the cognitive load theory and dual-system theory.Design/methodology/approachThe authors executed a controlled experiment and analyzed the results using a two-stage regression method that linked visual appeal, bookmarking and purchase intent. Further empirical analysis was conducted to authenticate the authors' proposed model, utilizing real-world mobile commerce data from a clothing company.FindingsThis study's findings suggest that visual appeal influences purchase intent primarily through the full mediation of bookmarking, rather than exerting a direct influence. Furthermore, an increase in colorfulness corresponds positively with visual appeal, while visual complexity exhibits an inverted U-shaped relationship with it.Originality/valueThis study provides novel insights into the choice-set formation process through the theoretical lens of dual-system theory. Additionally, the authors employed an image processing technique to quantify a product's visual appeal as depicted in a photograph. This study also incorporates a comprehensive econometric analysis to connect the objective aspects of visual appeal with subjective responses.

Predicting People’s Concentration and Movements in a Smart City

With the rapid growth of urbanization and the proliferation of mobile phone usage, smart city initiatives have gained momentum in leveraging data-driven insights to enhance urban planning and resource allocation. This paper proposes a novel approach for predicting people’s concentration and movements within a smart city environment using mobile phone data provided by telecommunication operators. By harnessing the vast amount of anonymized and aggregated mobile phone data, we present a predictive framework that offers valuable insights into urban dynamics. The methodology involves collecting and processing location-based data obtained from telecommunication operators. Using machine learning techniques, including clustering and spatiotemporal analysis, we developed models to identify patterns in people’s movements and concentration across various city regions. Our proposed approach considers factors such as time of day, day of the week, and special events to capture the intricate dynamics of urban activities. The predictive models presented in this paper demonstrate the ability to predict areas of high concentration of people, such as commercial districts during peak hours, as well as the people flow during the time. These insights have significant implications for urban planning, traffic management, and resource allocation. Our approach respects user privacy by working with aggregated and anonymized data, ensuring compliance with privacy regulations and ethical considerations. The proposed models were evaluated using real-world mobile phone data collected from a smart city environment in Lisbon, Portugal. The experimental results demonstrate the accuracy and effectiveness of our approach in predicting people’s movements and concentration. This paper contributes to the growing field of smart city research by providing a data-driven solution for enhancing urban planning and resource allocation strategies. As cities continue to evolve, leveraging mobile phone data from telecommunication operators can lead to more efficient and sustainable urban environments.

SUMMIT: A multi-modal agent-based co-simulation of urban public transport with applications in contingency planning

In this article, we present SUMMIT (Singapore Urban Multi-Modal Integrated Transport Simulator), a multi-modal agent-based simulation platform for public transport calibrated using real world mobility data sets from Singapore. SUMMIT uses a co-simulation approach in order to model multiple modes of public transport in tandem with commuters seamlessly transitioning between different mode simulations. This is implemented through a message passing framework codenamed Fabric (Fast, Agent-Based, Reproducible, Integrated Co-simulation) which helps to align and synchronise the simulation as it transitions through the different simulation time steps. Three stand-alone simulators, each modelling a key mode in the Singapore public transport system, namely train, bus and taxi, are integrated with a commuter control model, which simulates the commuter behaviour and takes care of commuters’ transfer between different modes. We present one application of SUMMIT in public transport contingency planning, where train services are disrupted in multiple lines of Singapore Mass Rapid Transit (MRT) network. Different scenarios are simulated and key performance indicators are compared to help planners evaluate the efficacy of mitigation measures. The results show that bridging bus services, a commonly used mitigation measure, are effective in reducing crowd sizes within the train stations. However, we demonstrate that bridging buses could also result in worsening of commuters’ travel time due to over-demand for these buses. Our results also show early dissemination of information to commuters during a train service disruption could reduce the negative impacts of the disruption event significantly.

PSARE: A RL-Based Online Participant Selection Scheme Incorporating Area Coverage Ratio and Degree in Mobile Crowdsensing

Mobile crowdsensing (MCS) is a cost-effective paradigm for gathering real-time and location-related urban sensing data. To complete MCS tasks, MCS platform needs to exploit the trajectory of participants (vehicles or individuals, etc.) for effectively choosing participants. On one hand, the existing works usually assume that platform has possessed the abundant historical movement trajectory for participant selection, or can accurately predict the movement of participant before selection, but this assumption is impractical for many MCS applications, for some candidates have just arrived without sufficient mobility profiles, so-called trajectory from-scratch, or cold-trajectory issue. On the other hand, most of works only considers the coverage ratio of the sensing area, while some hotspots should be sensed frequently, so-called coverage degree of hotspots. To solve the issue, this paper proposes a reinforcement learning (RL) based, i.e., an improved Q-learning based online participant selection scheme to incorporate both coverage ratio and degree, PSARE. First, to solve the explosion of state-value table in traditional tabular Q-learning, an improved two-level Q-learning method is proposed to select participants in online way so as to achieve high long-term return. Specifically, in each selection round, PSARE dynamically compresses all the real participants (RPs) into several virtual participants (VPs) using the available historical trajectories of RPs, and the VP-based state-value table is constructed and constantly updated (i.e., the first level). Then, after selecting the VP through looking up the table, PARSE chooses the RP with the largest expected reward in this VP using epsilon-greedy way to balance the effect of exploration and exploitation (i.e., the second level). Moreover, the reward function is designed to measure the MCS coverage quality, including both coverage degree of hotspots and coverage ratio of target area. Thorough experiments on real-world mobility data set demonstrate that PSARE outperforms than other RL based online participant selection schemes (including deep Q-learning network) and traditional offline selection methods.

Smartphone-Based Localization for Passengers Commuting in Traffic Hubs.

Passengers commute between different modes of transportation in traffic hubs, and passenger localization is a key component for the effective functioning of these spaces. The smartphone-based localization system presented in this work is based on the 3D step and heading approach, which is adapted depending on the position of the smartphone, i.e., held in the hand or in the front pocket of the trousers. We use the accelerometer, gyroscope and barometer embedded in the smartphone to detect the steps and the direction of movement of the passenger. To correct the accumulated error, we detect landmarks, particularly staircases and elevators. To test our localization algorithm, we have recorded real-world mobility data in a test station in Munich city center where we have ground truth points. We achieve a 3D position accuracy of 12 m for a smartphone held in the hand and 10 m when the phone is placed in the front pocket of the trousers.

An Evaluation Method of Road Link Functionality Using Individual Trajectory Data and Weighted Network Analysis

Road links within a city are hierarchical according to their structure and function. Upper-level road links, such as highways and arterials, are designed to maintain higher mobility and traffic flow, while lower-level road links should be more accessible. However, depending on the origin-destination demand pattern (O-D), drivers’ route choice, land use, and urban infrastructure, the actual usage pattern of roads could be different from the designed intention. This difference ultimately puts a load on certain road links and causes traffic jams. In order to handle this issue, it is necessary to create an appropriate evaluation method for the functionality of road links in advance. The research suggests an evaluation method to examine the functionalities of the roadways by using real-world mobility data and weighted network analysis. In the study, the roles of links were defined and quantified by three network attributes, in-strength, out-strength, and betweenness centrality. Derived attributes were used to cluster links with similar travel patterns. Furthermore, the concept of link reliability was introduced to measure the reliance of the network on individual links. Those network indices make it possible to evaluate the functioning of roads based on people’s travel patterns and to detect critical links that are irreplaceable and difficult to detour. This information can be used to determine the priorities of upcoming improvements and ultimately improve the efficiency of operation and maintenance of the road link networks.

Dynamic Origin-Destination Flow Prediction Using Spatial-Temporal Graph Convolution Network With Mobile Phone Data

Massive mobile phone data provide continuous and large-scale dynamic origin–destination (OD) flow information for multiple modes of transportation. In this study, we represent the dynamic OD flows obtained from mobile phone data as time-dependent graphs and propose two novel spatial-temporal graph convolutional network (STGCN)-based models to predict dynamic OD flows. Both models directly operate on the graph-structured OD flows, capture correlations among OD flows far apart in the Euclidean space, and fully explore the complex spatial-temporal features. We first formulate OD flows as explicit edges that specify the travels between two locations and propose an edge-focused STGCN. The edge-focused STGCN applies a novel three-step strategy to effectively update edge features in large-scale graphs. Second, we formulate OD flows as vertices in graph and propose a vertex-focused STGCN. The vertex-focused STGCN infers the relations among OD flows by establishing an adjacency matrix based on the temporal similarity between OD flows. The proposed models were validated using real-world mobile phone data collected in Kunshan, China. OD flows in the next hour were predicted, and the mean absolute percent errors of the edge-focused STGCN and the vertex-focused STGCN were 1.755% and 1.672%, respectively; both were significantly lower than the current baseline models.

Dosage Frequency Effects on Treatment Outcomes Following Self-managed Digital Therapy: Retrospective Cohort Study.

BackgroundAlthough the efficacy of high-dose speech-language therapy (SLT) for individuals with poststroke aphasia has been established in the literature, there is a gap in translating these research findings to clinical practice. Therefore, patients continue to receive suboptimal amounts of SLT, with negative consequences for their functional communication recovery. Recent research has identified self-managed digital health technology as one way to close the dosage gap by enabling high-intensity therapy unrestricted by clinician availability or other practical constraints. However, there is limited empirical evidence available to rehabilitation professionals to guide dose prescriptions for self-managed SLT despite their increasing use in the COVID-19 era and likely beyond.ObjectiveThis study aims to leverage real-world mobile health data to investigate the effects of varied dosage frequency on performance outcomes for individuals with poststroke speech, language, and cognitive deficits following a 10-week period of self-managed treatment via a commercially available digital health platform.MethodsAnonymized data from 2249 poststroke survivors who used the Constant Therapy app between late 2016 and 2019 were analyzed. The data included therapy tasks spanning 13 different language and cognitive skill domains. For each patient, the weekly therapy dosage was calculated based on the median number of days per week of app use over the 10-week therapy period, binned into groups of 1, 2, 3, 4, or ≥5 days per week. Linear mixed-effects models were run to examine change in performance over time as a function of dosage group, with post hoc comparisons of slopes to evaluate the performance gain associated with each additional day of practice.ResultsAcross all skill domains, linear mixed-effects model results showed that performance improvement was significantly greater for patients who practiced 2 (β=.001; t15,355=2.37; P=.02), 3 (β=.003; t9738=5.21; P<.001), 4 (β=.005; t9289=7.82; P<.001), or ≥5 (β=.005; t6343=8.14; P<.001) days per week compared with those who only practiced for 1 day per week. Post hoc comparisons confirmed an incremental dosage effect accumulating with each day of practice (ie, 1 day vs 2 days, 2 days vs 3 days, and 3 days vs 4 days), apart from 4 days versus ≥5 days of practice per week. The result of greater improvement for higher versus lower dosage frequency groups was true not only across all domains but also within a majority of individual subdomains.ConclusionsThe findings from this study demonstrated that increased dosage frequency is associated with greater therapy gains over a 10-week treatment period of self-managed digital therapy. The use of real-world data maximizes the ecological validity of study results and makes the findings more generalizable to clinical settings. This study represents an important step toward the development of optimal dose recommendations for self-managed SLT.

Toward Accurate Spatiotemporal COVID-19 Risk Scores Using High-Resolution Real-World Mobility Data

As countries look toward re-opening of economic activities amidst the ongoing COVID-19 pandemic, ensuring public health has been challenging. While contact tracing only aims to track past activities of infected users, one path to safe reopening is to develop reliable spatiotemporal risk scores to indicate the propensity of the disease. Existing works which aim at developing risk scores either rely on compartmental model-based reproduction numbers (which assume uniform population mixing) or develop coarse-grain spatial scores based on reproduction number (R0) and macro-level density-based mobility statistics. Instead, in this article, we develop a Hawkes process-based technique to assign relatively fine-grain spatial and temporal risk scores by leveraging high-resolution mobility data based on cell-phone originated location signals. While COVID-19 risk scores also depend on a number of factors specific to an individual, including demography and existing medical conditions, the primary mode of disease transmission is via physical proximity and contact. Therefore, we focus on developing risk scores based on location density and mobility behaviour. We demonstrate the efficacy of the developed risk scores via simulation based on real-world mobility data. Our results show that fine-grain spatiotemporal risk scores based on high-resolution mobility data can provide useful insights and facilitate safe re-opening.

Unsupervised Learning for Human Mobility Behaviors

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.

Detecting variations in ovulation and menstruation during the COVID-19 pandemic, using real-world mobile app data.

As war and famine are population level stressors that have been historically linked to menstrual cycle abnormalities, we hypothesized that the COVID-19 pandemic could similarly affect ovulation and menstruation among women. We conducted a retrospective cohort study examining changes in ovulation and menstruation among women using the Natural Cycles mobile tracking app. We compared de-identified cycle data from March-September 2019 (pre-pandemic) versus March-September 2020 (during pandemic) to determine differences in the proportion of users experiencing anovulation, abnormal cycle length, and prolonged menses, as well as population level changes in these parameters, while controlling for user-reported stress during the pandemic. We analyzed data from 214,426 cycles from 18,076 app users, primarily from Great Britain (29.3%) and the United States (22.6%). The average user was 33 years of age; most held at least a university degree (79.9%). Nearly half (45.4%) reported more pandemic-related stress. Changes in average cycle and menstruation lengths were not clinically significant, remaining at 29 and 4 days, respectively. Approximately 7.7% and 19.5% of users recorded more anovulatory cycles and abnormal cycle lengths during the pandemic, respectively. Contrary to expectation, 9.6% and 19.6% recorded fewer anovulatory cycles and abnormal cycle lengths, respectively. Women self-reporting more (32.0%) and markedly more (13.6%) stress during the pandemic were not more likely to experience cycle abnormalities. The COVD-19 pandemic did not induce population-level changes to ovulation and menstruation among women using a mobile app to track menstrual cycles and predict ovulation. While some women experienced abnormalities during the pandemic, this proportion was smaller than that observed prior to the pandemic. As most app users in this study were well-educated women over the age of 30 years, and from high-income countries, their experience of the COVID-19 pandemic might differ in ways that limit the generalizability of these findings.

Vector-based pedestrian navigation in cities.

How do pedestrians choose their paths within city street networks? Researchers have tried to shed light on this matter through strictly controlled experiments, but an ultimate answer based on real-world mobility data is still lacking. Here, we analyze salient features of human path planning through a statistical analysis of a massive dataset of GPS traces, which reveals that (1) people increasingly deviate from the shortest path when the distance between origin and destination increases and (2) chosen paths are statistically different when origin and destination are swapped. We posit that direction to goal is a main driver of path planning and develop a vector-based navigation model; the resulting trajectories, which we have termed pointiest paths, are a statistically better predictor of human paths than a model based on minimizing distance with stochastic effects. Our findings generalize across two major US cities with different street networks, hinting to the fact that vector-based navigation might be a universal property of human path planning.

Hyperbolic K-means for traffic-aware clustering in cloud and virtualized RANs

As the internet and connected objects gain more and more in popularity, serving the ever-increasing data traffic becomes a challenge for the mobile operators. The traditional cellular radio access network (RAN), where each base station is co-located with its own processing unit and is responsible for a specific geographic area, has evolved first with the so-called Cloud RAN (C-RAN), and is currently undergoing further architectural evolution under the virtualized RAN (vRAN), Software-Defined RAN (SD-RAN), and Open RAN (O-RAN) architectures. In all these versions, the data processing units can be dynamically centralized into a pool and shared between several base stations, enlarging the geographical view for scheduling and resource allocation algorithms. For instance, resource utilization is improved by avoiding resource idling during off-peak hours. C-RAN and vRAN gains depend strongly on the clustering scheme of radio units (RRHs and RUs). In this paper, we propose a novel radio clustering algorithm that takes into account both the traffic demand and the position of stations, by using the hyperbolic distance in 3-dimensions. We introduce a modified K-means clustering algorithm, called Hyperbolic K-means, and show that this generates geographically compact RU clusters with traffic charge equally shared among them. Application of our algorithm on real-world mobile data traffic, collected from the cities of Nantes and Lille in France, shows an increase in resource utilization by 25%, and a reduction in deployment cost by 15%, compared to the standard RAN. Furthermore, the performance of our Hyperbolic K-means algorithm is compared extensively against alternative C-RAN clustering proposals from the literature and is shown to outperform them, in resource utilization as well as in cost reduction.

Real-World Mobility and Environmental Data for the Assessment of In-Vehicle Battery Capacity Fade

This work develops scenario-based analyses for predicting in-vehicle performance degradation of automotive traction batteries. It combines recent capacity performance-based models of NCM-LMO Li-ion (Nickel Cobalt Manganese Oxide—Lithium Manganese Oxide) variant batteries with real-world vehicle driving data from different geographical areas of Europe. The analysis addresses different battery and vehicle architectures (PHEVs (Plug-in Hybrid Electric Vehicles) and BEVs (Battery Electric Vehicles)) combined with different recharging strategies and mobility patterns and environmental temperatures. The mobility pattern datasets used in this analysis refer to six European cities and include up to 508,609 private vehicles, corresponding to 1.78 billion GPS records, 9.1 million trips and parking events and a total driven distance of 106.1 million kilometers. The results show the effect that the environmental temperature, the recharging power, and the driven kilometers have on the calendar and cycling aging. The majority of the combinations of the considered vehicle architectures and recharge strategies do not lead to battery capacity drop below 80% of its nominal value in less than five calendar years for a usage profile of up to 1000 km/month.

Toward Discriminating and Synthesizing Motion Traces Using Deep Probabilistic Generative Models.

Mining knowledge from human mobility, such as discriminating motion traces left by different anonymous users, also known as the trajectory-user linking (TUL) problem, is an important task in many applications requiring location-based services (LBSs). However, it inevitably raises an issue that may be aggravated by TUL, i.e., how to defend against location attacks (e.g., deanonymization and location recovery). In this work, we present a Semisupervised Trajectory- User Linking model with Interpretable representation and Gaussian mixture prior (STULIG)-a novel deep probabilistic framework for jointly learning disentangled representation of user trajectories in a semisupervised manner and tackling the location recovery problem. STULIG characterizes multiple latent aspects of human trajectories and their labels into separate latent variables, which can be then used to interpret user check-in styles and improve the performance of trace classification. It can also generate synthetic yet plausible trajectories, thus protecting users' actual locations while preserving the meaningful mobility information for various machine learning tasks. We analyze and evaluate STULIG's ability to learn disentangled representations, discriminating human traces and generating realistic motions on several real-world mobility data sets. As demonstrated by extensive experimental evaluations, in addition to outperforming the state-of-the-art methods, our method provides intuitive explanations of the classification and generation and sheds lights on the interpretable mobility mining.

Joint Container Placement and Task Provisioning in Dynamic Fog Computing

Fog computing has emerged as a promising technology that can bring cloud applications closer to the devices at the network edge. The fog infrastructure contains mainly distributed and heterogeneous fog devices such as in the context of the Internet of Things. Unlike traditional data centers, those devices are characterized by sporadic resources availability, mobility, and increased flexibility. However, resource allocation mechanisms proposed currently for fog computing still lack the support of dynamic behavior. In this article, we propose novel resource management algorithms capable of flexible service provisioning in a dynamic fog computing environment. Specifically, the joint problem of container placement and task provisioning is formulated with integer linear programming. Due to its NP-hardness, we propose a low-complex particle-swarm-optimization-based metaheuristic and a greedy heuristic. Our solutions aim to optimize the number of served end-users with a predefined delay-threshold while considering dynamic fog nodes behavior/mobility and resources availability of fog nodes. Using real-world mobility data sets and different resources’ availability models, conducted simulations demonstrate that the PSO-based algorithm achieves near-optimal results. Whereas, the greedy algorithm realizes only 10%–30% less success ratio than the optimal solution with negligible execution time.

Inference of node attributes from social network assortativity

Social networks are known to be assortative with respect to many attributes, such as age, weight, wealth, level of education, ethnicity and gender: Similar people according to these attributes tend to be more connected. This can be explained by influences and homophily. Independently of its origin, this assortativity gives us information about each node given its neighbors. Assortativity can thus be used to improve individual predictions in a broad range of situations, when data are missing or inaccurate. This paper presents a general framework based on probabilistic graphical models to exploit social network structures for improving individual predictions of node attributes. Using this framework, we quantify the assortativity range leading to an accuracy gain in several situations, with various individual prediction profiles. We finally show how specific characteristics of the network can enhance performances further. For instance, the gender assortativity in real-world mobile phone data drastically changes according to some communication attributes. In this case, using the network topology indeed improves local predictions of node labels and moreover enables inferring missing node labels based on a subset of known vertices. In both cases, the performances of the proposed method are statistically significantly superior to the ones achieved by state-of-the-art label propagation and feature extraction schemes in most settings.

Optimizing utilization in cellular radio networks using mobility data

The main resource for any telecom operator is the physical radio cell network. We present two related methods for optimizing utilization in radio networks: Tetris optimization and selective cell expansion. Tetris optimization tries to find the mix of users from different market segments that provides the most even load in the network. Selective cell expansion identifies hotspot cells, expands the capacity of these radio cells, and calculates how many subscribers the radio network can handle after the expansions. Both methods are based on linear programming and use mobility data, i.e., data defining where different categories of subscribers tend to be during different times of the week. Based on real-world mobility data from a region in Sweden, we show that Tetris optimization based on six user segments made it possible to increase the number of subscribers by 58% without upgrading the physical infrastructure. The same data show that by selectively expanding less than 6% of the cells we are able to increase the number of subscribers by more than a factor of three without overloading the network. We also investigate the best way to combine Tetris optimization and selective cell expansion.

Fine-Grained Multitask Allocation for Participatory Sensing With a Shared Budget

For participatory sensing, task allocation is a crucial research problem that embodies a tradeoff between sensing quality and cost. An organizer usually publishes and manages multiple tasks utilizing one shared budget. Allocating multiple tasks to participants, with the objective of maximizing the overall data quality under the shared budget constraint, is an emerging and important research problem. We propose a fine-grained multitask allocation framework (MTPS), which assigns a subset of tasks to each participant in each cycle. Specifically, considering the user burden of switching among varying sensing tasks, MTPS operates on an attention-compensated incentive model where, in addition to the incentive paid for each specific sensing task, an extra compensation is paid to each participant if s/he is assigned with more than one task type. Additionally, based on the prediction of the participants' mobility pattern, MTPS adopts an iterative greedy process to achieve a near-optimal allocation solution. Extensive evaluation based on real-world mobility data shows that our approach outperforms the baseline methods, and theoretical analysis proves that it has a good approximation bound.