14th Jun, 2021
Dhofar University
Question
Asked 3rd May, 2016
Can anyone describe some ways in which scholars of mobility use Big Data to capture and analyze cross-border human mobility in time and space?
More especifically how we could use Big Data to capture and analyze cross-border human mobility in time and space?
Most recent answer
1 Recommendation
Popular answers (1)
I am assuming that you are interested in the methodological aspect of the topic rather than specific tools.
You can use geo-located tweets or cellphone call detail records to capture the mobility at the individual level. The availability of these two data types depends on the geographic area of interest. While you can get a wealth of data in developed/developing countries, chances are low when it comes to countries with civil war (e.g., Syria). You can still estimate cross-border mobility using approaches like the gravity model or the radiation model.
When it comes to the analyses, you can do some time series analysis assuming that you capture the cross-border mobility over time. This is just an example, eventually your analyses depend on your research question. If you provide more specific modeling case, you may get more tailored answers.
4 Recommendations
All Answers (9)
Hi,
It totally depends on the data set and tools that is used. I would mostly prefer a hadoop framework for map-reduce (java/python) and a live connection to Tableau visualization for temporal and spatial analysis. I am currently working on a similar topic and this works.
2 Recommendations
Thank you. I think Tableau visualization offers the ability to create different views of our data and change them, as our needs evolve. Tableau also allows us to filter or create another views of our data.
2 Recommendations
I am assuming that you are interested in the methodological aspect of the topic rather than specific tools.
You can use geo-located tweets or cellphone call detail records to capture the mobility at the individual level. The availability of these two data types depends on the geographic area of interest. While you can get a wealth of data in developed/developing countries, chances are low when it comes to countries with civil war (e.g., Syria). You can still estimate cross-border mobility using approaches like the gravity model or the radiation model.
When it comes to the analyses, you can do some time series analysis assuming that you capture the cross-border mobility over time. This is just an example, eventually your analyses depend on your research question. If you provide more specific modeling case, you may get more tailored answers.
4 Recommendations
Thank you very much. You are right, I am interested in the methodological aspect of the topic. Your answer is very useful for me!
3 Recommendations
Hi Silvia,
In the analysis of human mobility, there are two choices you have to take. First, the choice of the dataset:
- mobile phone data are widely used to study human mobility, and most of the mobility measures used in literature are defined over mobile phone data. Generally mobile phone data aren't usually publicly available, but recently a sample of mobile phone data have been provided by Orange for the D4D challenge (see this link: http://www.d4d.orange.com/en/Accueil). For a discussion on the pros and cons of using mobile phone data for human mobility analysis I suggest you the following papers:
- Returners and Explorers dichotomy in Human Mobility, Nature Communications, 2015;
- Understanding individual human mobility patterns, Nature 2008;
- Measures of Human Mobility Using Mobile Phone Records Enhanced with GIS Data, PlosOne, 2015.
- GPS data are another widely used proxy to observe human mobility. In particular, several studies analyzed traces produced by GPS devices installed on cars. As mobile phone data, GPS traces from cars are not publicly available. You can find a discussion on the usage of GPS data for the analysis on human mobility in the following papers:
- Understanding the patterns of car travel, EPJ Special Topics 2013
- Returners and Explorers dichotomy in Human Mobility, Nature Communications 2015.
- Other proxies for human mobility are georeferenced tweets and data from location-based social networks like Foursquare. You can find a free Foursquare dataset at this link: https://archive.org/details/201309_foursquare_dataset_umn
- You can produce synthetic data using an individual mobility model or a migration model. Widely used individual mobility models are the EPR model (Modelling the scaling properties of human mobility, Nature Physics, 2010) and the more recent d-EPR model (Returners and Explorers dichotomy in Human Mobility, Nature Communications, 2015, and Human Mobility Modelling: exploration and preferential return meet the gravity model, 2016). Widely used migration models are the gravity model and the radiation model (A universal model for mobility and migration patterns, Nature, 2012).
The second choice you have to do is the mobility measure/model to use. You can use Origin-Destination matrices if you want to study the network of movements between two locations (Discovering the Geographical Borders of Human Mobility, 2012), the radius of gyration if you want to estimate the characteristic distance traveled by individuals (Understanding the patters of car travel, 2013 or Returners and Explorers dichotomy in Human Mobility, 2015), the mobility entropy if you want to estimate how predictable individuals' movements are (Limits of predictability in human mobility, Science, 2010 or Using mobile phone data to study the link between human mobility and socio-economic development, 2015).
To study cross-border mobility (I suppose you mean mobility between different countries), you need mobility data covering several countries. Generally, mobile phone data and GPS data are provided for a single country, so they do not fit completely your purpose. You can use georeferenced tweets or Foursquare data: they are less detailed than mobile phone of GPS data but they are available for every country.
If you want to study migration patterns between countries, you can use the International Migration database (https://stats.oecd.org/Index.aspx?DataSetCode=MIG) and refer to the following website and the related Science paper: http://www.global-migration.info/
Conference Paper Human Mobility Modelling: Exploration and Preferential Retur...
3 Recommendations
Thank you very much for your complete answer. I will read the articles and then I will utilise the information!You are an expert in the field! Many thanks for your help!
2 Recommendations
I've just came across this linked article while searching the literature for my project. It seems to be useful for addressing your question.
2 Recommendations
Similar questions and discussions
I need the D4D Challenge dataset (CDR data). Anybody has already downloaded it?
We want to do research on CDR data. The best data set for this purpose is D4D challenge data set (http://www.d4d.orange.com/en/Accueil). Unfortunately the challenge is closed and the data set is no longer available. Anyone who has already downloaded it can please provide me a private link or something?
Does anyone know any other public CDR dataset?
Thanks
Related Publications
In this chapter, we estimate human mobility between countries worldwide on the basis of global statistics on tourism and air passenger traffic. Adjusting and merging the data from these two sources through a simple set of procedures enabled us to counter some of their individual limitations. The resulting open-access dataset, which covers more than...
Covid-19 interventions are greatly affecting patterns of human mobility. Changes in mobility during Covid-19 have differed across socio-economic gradients during the first wave. We use fine-scale network mobility data in Ontario, Canada to study the association between three different mobility measures and four socio-economic indicators throughout...
