Blog

How our Data Scientists Work with Academia

Alex Rutherford
Mar 26, 2014

In my role as a data scientist at Global Pulse, working in the field of big data for public good, it's important for me to exchange ideas with the academic community and gain fresh knowledge from them. Our research team across three Pulse Labs (New York, Kampala, Jakarta) is working on ways to engage, ranging from formal to more informal modes of collaboration. I'll talk about some of those ways we interact with researchers and academics in this post.

We meet smart people at scientific conferences

At Global Pulse we have to be up to date with the latest methods and ideas in academic research. We also want to encourage early career scientists to see not only the reward in using their Big Data expertise for development purposes, but also the unique intellectual challenges that these pose. For this reason, you may see some me and my colleagues at scientific conferences such as the recent Sentiment Analysis Symposium, the Workshop on Information in Networks or the ICTD Conference. Some of our best collaborations have emerged from interactions at conferences.
 
In fact, there is a growing trend within the machine learning, computational social science and other communities, to promote applications of research to development and humanitarian problems. In recent years conferences such as IEEE Global Humanitarian Technology Conference and HumNet have recognised this fact. This year the prestigious computer science conference Knowledge Discovery in Data has Data Mining for Social Good as a key theme. Global Pulse was also pleased to be involved in the 2013 NetMob conference - the main conference for the scientific analysis of mobile datasets. 2013 was the first year that telecommunications company Orange made a subset of Call Detail Records from Cote d’Ivoire available for researchers to use, and winning research projects were showcased on a special day at the NetMob conference.
 

We work with academics in their home institutions

We are also working directly with academics in their home institutions. For example, in collaboration with researchers from Telefonica Research and PhD students from Universidad Polit├ęcnica de Madrid's specialist Technology and Innovation for Development Centre we have been working on a project to analyze anonymized mobile data to understand population displacements following floods. I'll share more details about that project on this website, as our research progresses. 
 
In 2013 our Pulse Lab New York research team, together with colleagues from UNDP, supervised a student project for a Data Science course at Columbia University. The project investigated the potential of analyzing text from a large digital corpora of news articles as a way to predict or anticipate rising tension, conflict or unrest. This involved extracting relationships and sentiments between different actors and countries in each country to find changing trends which might not be obvious from examining articles in isolation.
 
Currently our research team is supervising a project for students of a visualisation course at NYU Polytechnic School of Engineering. These students are helping to build an open source library for visualising global trade flows  -and that can be expanded and modified to fit our future research programs.
 
Pulse Lab Kampala (PLK) is also reaching out, and has links to Makerere University in Uganda, where - for example - researchers in the AI-DEV group have been investigating the use of artificial intelligence methods for gathering information in the developing world
 

We Host Research Fellows in our Pulse Labs

At the more in-depth end of our engagement with academia, is our Fellowship Program for current PhD students or post-doctoral researchers. Fellows come and work with us on their own research question in our Headquarters in New York or in one of our labs in Jakarta or Kampala. Fellowship and internship positions open periodically, and are posted on our vacancies page.
 
To date, we have hosted fellows from Stockholm University, Leuven University College in Belgium among others in our New York lab.  Pulse Lab Jakarta (PLJ) has just taken on their first fellow from the Korea Advanced Institute of Science and Technology. These students have taken time from their studies to work on data analysis or a visualisation tool that feed directly into our research programs.  
 

The future is bright

What is very encouraging from all these interactions with young researchers is that they see very clearly the potential application of their skills to global development and humanitarian issues and are very motivated to do so. 
 
It seems that some progress has been made since Jeff Hammerbacher made his famous pronouncement that: 'the best minds of my generation are thinking about how to make people click ads… That sucks.'
 
Finally, this week we have started tweeting as the Global Pulse Data Science team at @UNGPDataScience so share your Big Data for Development (BD4D) research with us and join the conversation. 
 
Alex Rutherford is a data scientist at Pulse Lab New York
 

Add comment