Population estimation from mobile network traffic metadata

Khodabandelou, Ghazaleh; Gauthier, Vincent; Yacoubi, Mounîm El; Fiore, Marco

doi:10.1109/wowmom.2016.7523554

Cited by 36 publications

(26 citation statements)

References 15 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The approach in [15] mimics several of our solutions, as originally presented in [24]. It leverages regressions on the power-law model of population density in (2), where it uses the number of subscribers at 7 am as the σ i variable: this is semantically similar, but not identical, to the subscriber presence we employ as σ i .…”

Section: Dynamic Populationmentioning

confidence: 99%

Estimation of Static and Dynamic Urban Populations with Mobile Network Metadata

Khodabandelou

Gauthier

Fiore³

et al. 2019

IEEE Trans. on Mobile Comput.

Self Cite

View full text Add to dashboard Cite

Communication-enabled devices routinely carried by individuals have become pervasive, opening unprecedented opportunities for collecting digital metadata about the mobility of large populations. In this paper, we propose a novel methodology for the estimation of people density at metropolitan scales, using subscriber presence metadata collected by a mobile operator. Our approach suits the estimation of static population densities, i.e., of the distribution of dwelling units per urban area contained in traditional censuses. More importantly, it enables the estimation of dynamic population densities, i.e., the time-varying distributions of people in a conurbation. By leveraging substantial real-world mobile network metadata and ground-truth information, we demonstrate that the accuracy of our solution is superior to that granted by state-of-the-art methods in practical heterogeneous urban scenarios.

show abstract

Section: Dynamic Populationmentioning

confidence: 99%

Estimation of Static and Dynamic Urban Populations with Mobile Network Metadata

Khodabandelou

Gauthier

Fiore³

et al. 2019

IEEE Trans. on Mobile Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Researchers have established connections between the underlying population distribution obtained from census estimates with telecommunications data [19]. A recent study demonstrated that "presence" information (i.e., an estimate of the number of users in a certain space based on their last interaction with the mobile network) can provide better correlation with census data [7]. The same study also proposed a method to evaluate real-time population distribution and showed reasonable results, although ground truth was not available.…”

Section: B Mobile Phone Databases For Population Estimationmentioning

confidence: 99%

“…Khodabandelou et al [7] proposed the use of mobile activity and census data to provide a real-time population estimate. A model was established using overnight values and census data.…”

Section: B Mobile Phone Activity Data Analysismentioning

confidence: 99%

“…Internet activity, however, shows a significantly different map. In fact, Internet data is not considered a good proxy for human activity due to Internet traffic generated by applications running in the background [7]. **San Siro Stadium is partially covered, which poses additional constraints to an emergency landing, but it is used as a general example.…”

Section: Donato and Atkinsmentioning

confidence: 99%

“…The first contribution of the paper is to explore an emerging information source, the mobile phone activity database, for occupancy estimation. This idea has received recent attention outside aerospace [6][7][8][9] but to date has not been applied to aviation.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Evaluating Risk to People and Property for Aircraft Emergency Landing Planning

Donato

Atkins

2017

Journal of Aerospace Information Systems

View full text Add to dashboard Cite

General aviation and small unmanned aircraft systems are less redundant, may be less thoroughly tested, and are flown at lower cruise altitudes than commercial aviation counterparts. These factors result in a higher probability of a forced or emergency landing scenario. Currently, general aviation relies on the pilot to select a landing site and plan a trajectory, even though workload in an emergency is typically high, and decisions must be made rapidly. Although sensors can provide local real-time information, awareness of more distant or occluded regions requires database and/or offboard data sources. This paper considers different data sources and how to process these data to inform an emergency landing planner regarding risks posed to property, people on the ground, and the aircraft itself. Detailed terrain data are used for selection of candidate emergency landing sites. Mobile phone activity is evaluated as a means of real-time occupancy estimation. Occupancy estimates are combined with population census data to estimate emergency landing risk to people on the ground. Openly available databases are identified and mined as part of an emergency landing planning case study. = total number of country codes n g = total number of grid cells n t = total number of time intervals R a = risk based on landing area R h = risk to people on ground R ls = overall landing site risk R p = risk to property on ground R v = risk to vehicle t = time, min t 10 m = starting time of 10 min time interval, min W h , W p , W v , W a = weights associated with landing site risks w = A transition cost weight Θ = median of maximum aggregated SMS and call activity during a day, for a day of the week λ census = occupancy from census datâ λ census = normalized occupancy from census data λ rt = real-time occupancy estimation λ 0 = A fixed transition cost element Φ = median of aggregated SMS and call activity for each grid cell, day of the week and time interval of the day Φ grid = sum of Φ for all grid cellŝ Φ = normalized Φ ϕ = aggregated SMS and call activitŷ ϕ = normalized ϕ ϕ 1 = activity in terms of SMS-in ϕ 2 = activity in terms of SMS-out ϕ 3 = activity in terms of call-in ϕ 4 = activity in terms of call-out ϕ 5 = activity in terms of internet ϕ i = modified activity feature for heat map plot

show abstract

Combining Bayesian Inference and Clustering for Transport Mode Detection from Sparse and Noisy Geolocation Data

Bachir

Khodabandelou

Gauthier

et al. 2019

Machine Learning and Knowledge Discovery in Databases

Self Cite

View full text Add to dashboard Cite

Large-scale and real-time transport mode detection is an open challenge for smart transport research. Although massive mobility data is collected from smartphones, mining mobile network geolocation is non-trivial as it is a sparse, coarse and noisy data for which real transport labels are unknown. In this study, we process billions of Call Detail Records from the Greater Paris and present the first method for transport mode detection of any traveling device. Cellphones trajectories, which are anonymized and aggregated, are constructed as sequences of visited locations, called sectors. Clustering and Bayesian inference are combined to estimate transport probabilities for each trajectory. First, we apply clustering on sectors. Features are constructed using spatial information from mobile networks and transport networks. Then, we extract a subset of 15% sectors, having road and rail labels (e.g., train stations), while remaining sectors are multi-modal. The proportion of labels per cluster is used to calculate transport probabilities given each visited sector. Thus, with Bayesian inference, each record updates the transport probability of the trajectory, without requiring the exact itinerary. For validation, we use the travel survey to compare daily average trips per user. With Pearson correlations reaching 0.96 for road and rail trips, the model appears performant and robust to noise and sparsity.

show abstract

Population estimation from mobile network traffic metadata

Cited by 36 publications

References 15 publications

Estimation of Static and Dynamic Urban Populations with Mobile Network Metadata

Estimation of Static and Dynamic Urban Populations with Mobile Network Metadata

Evaluating Risk to People and Property for Aircraft Emergency Landing Planning

Combining Bayesian Inference and Clustering for Transport Mode Detection from Sparse and Noisy Geolocation Data

Contact Info

Product

Resources

About