A5026620476- Human Mobility and Location-Based Analysis
- Urban Transport and Accessibility
- Urban Design and Spatial Analysis
- Land Use and Ecosystem Services
- Impact of Light on Environment and Health
- ICT in Developing Communities
- Complex Network Analysis Techniques
- Energy Load and Power Forecasting
- Explainable Artificial Intelligence (XAI)
- Spatial and Panel Data Analysis
- Theoretical and Computational Physics
- Complex Systems and Time Series Analysis
- Housing Market and Economics
- Remote Sensing and Land Use
- Machine Learning and Data Classification
- Image and Signal Denoising Methods
- Regional Economics and Spatial Analysis
- Chaos control and synchronization
- Medical Image Segmentation Techniques
- Insurance, Mortality, Demography, Risk Management
- COVID-19 epidemiological studies
- demographic modeling and climate adaptation
- Image Retrieval and Classification Techniques
- Regional Economic and Spatial Analysis
The Alan Turing Institute
2023
Singapore Institute of Technology
2020-2021
Orange (France)
2020-2021
University College London
2016-2020
École Normale Supérieure Paris-Saclay
2011
Centre National de la Recherche Scientifique
2011
Various methods have been developed independently to study the multifractality of measures in many different contexts. Although they all convey same intuitive idea giving a "dimension" sets where quantity scales similarly within space, are not necessarily equivalent on more rigorous level. This review article aims at unifying multifractal methodology by presenting theoretical framework and principal practical methods, namely moment method, histogram detrended fluctuation analysis (MDFA)...
Multifractal analysis offers a number of advantages to measure spatial economic segregation and inequality, as it is free categories boundaries definition problems insensitive some shape-preserving changes in the variable distribution. We use two datasets describing Kyoto land prices 1912 2012 derive city models from this data show that multifractal suitable describe heterogeneity prices. found particular sharp decrease multifractality, characteristic homogenisation, between older present...
High quality census data are not always available in developing countries. Instead, mobile phone becoming a popular proxy to evaluate the density, activity and social characteristics of population. They offer additional advantages: they updated real-time, include mobility information record visitors’ activity. However, we show with example Senegal that direct correlation between average both population density nighttime lights intensity may be insufficiently high provide an accurate...
Reliable and affordable access to electricity has become one of the basic needs for humans is, as such, at top development agenda. It contributes socio-economic by transforming whole spectrum people’s lives—food, education, healthcare. spurs new economic opportunities, thus improving livelihoods. Using a comprehensive dataset pseudonymized mobile phone records, we analyse impact electrification on attractiveness rural areas in Senegal. We extract communication mobility flows from call detail...
The Synthetic Population Catalyst (SPC) is an open-source tool for the simulation of populations. Building on previous efforts, synthetic populations can be created any area in England, from a small geographical unit to entire country, and linked geolocalised daily activities. In contrast most transport models, output focussed population itself way people socially interact together, rather than precise modelling volume trips one another. SPC therefore particularly well suited, example, study...
High quality census data are not always available in developing countries. Instead, mobile phone becoming a trending proxy to evaluate population density, activity and social characteristics. They offer additional advantages for infrastructure planning such as being updated real-time, including mobility information recording visitors' activity. We combine various sets from Senegal data's potential replace insufficient As an applied case, we test their ability at predicting domestic...
The interaction of all mobile species with their environment hinges on movement patterns: the places they visit and how frequently go there. In human society, where prevalent form cohabitation is in cities, highly dynamic diverse people fundamental to almost every aspect socio-economic life, including social interactions or disease spreading, ultimately key evolution urban infrastructure, productivity, innovation technology. However, despite crucial role spatio-temporal structure laws that...
Entropy relates the fast, microscopic behaviour of elements in a system to its slow, macroscopic state. We propose use it explain how, as complexity theory suggests, small scale decisions individuals form cities. For this, we offer first interpretation entropy for cities that reflects interactions between different places through interdependently linked states multiscale approach. With simulated patterns show structural spatial systems can be most probable configuration if interact across...
A reliable and affordable access to electricity has become one of the basic needs for humans is, as such, at top development agenda. It contributes socio-economic by transforming whole spectrum people's lives - food, education, health care; it spurs new economic opportunities thus improves livelihoods. Using a comprehensive dataset pseudonymised mobile phone records, provided market share leader, we analyse impact electrification on attractiveness rural areas in Senegal. We extract...
Multiscale signal analysis has been used since the early 1990s as a powerful tool for image processing, notably in linear case. However, nonlinear PDEs and associated operators have advantages over operators, preserving important features such edges images. In this paper, we focus on Hamilton-Jacobi defined with adaptive speeds or, alternatively, morphological fiters also called semi-flat operators. Semi-flat morphology were instroduced by H. Heijmans studied only case where speed (or...
Explainable artificial intelligence (XAI) provides explanations for not interpretable machine learning (ML) models. While many technical approaches exist, there is a lack of validation these techniques on real-world datasets. In this work, we present use-case XAI: an ML model which trained to estimate electrification rates based mobile phone data in Senegal. The originate from the Data Development challenge by Orange 2014/15. We apply two model-agnostic, local explanation and find that while...