12/19/2016

Linardi and Li, Ford Institute Faculty Research Grant

 

Assistant Professor Sera Linardi is one of two inaugural recipients of a Ford Institute Faculty Research Grant, a competitive grant program designed to encourage junior faculty to engage in human security research and writing. Combining violence data from UN peace keepers’ weekly logs in  Côte d’Ivoire with daily antenna transmission data from Orange Telecom, Professor Linardi and her colleagues have shown a pattern of increased call volume, more within-network calls, and shorter calls in the days preceding violent events. This research holds great potential to breakthrough our current limited understanding of local level violence, to better address its insidious effect on inter-group relations and potential escalation into national-level problems.

Assistant Professor Sera Linardi is one of two inaugural recipients of a Ford Institute Faculty Research Grant, a competitive grant program designed to encourage junior faculty to engage in human security research and writing. Combining violence data from UN peace keepers’ weekly logs in  Côte d’Ivoire with daily antenna transmission data from Orange Telecom, Professor Linardi and her colleagues have shown a pattern of increased call volume, more within-network calls, and shorter calls in the days preceding violent events (To read paper, click here). This research holds great potential to breakthrough our current limited understanding of local level violence, to better address its insidious effect on inter-group relations and potential escalation into national-level problems.

In order to test the potential of mobile communication data to improve prediction of violence at high temporal and spatial resolution, Dr. Linardi utilized the grant to hire Master’s student Lujing Li, of the School of Information Sciences. The team wrote an R library that makes it easy to access R’s machine learning functionality for prediction and to plot the output on GIS maps.

Linardi intends to use the library to better understand of how various techniques and variables for violence prediction, that are right now spread across multiple papers and context, work together in a single setting. During a recent presentation, Linardi and Li debuted the library, which will be made publicly available soon as the eventPrediction R package. 

To read paper, click here