Date of Award
1-11-2021
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Information Science
First Advisor
Nitin Agarwal
Abstract
The world-wide refugee problem has a long history, but continues to this day, and will unfortunately continue into the foreseeable future. Efforts to adequately anticipate, mitigate and prepare for refugee flows, however, are still lacking. There are many potential causes for refugee flows, but the published research has primarily focused on identifying ways to integrate already existing refugees into the various communities wherein they ultimately reside after having fled their home countries, rather than on the idea of figuring out ways to prevent the situations that initially caused them to flee. The model proposed in this dissertation uses a set of metrics that can be divided into three basic categories: 1) sociocultural, 2) socioeconomic, and 3) economic, which refers to the nature of each predictive feature. Each of these proposed predictive features are universal, in that they exist within each country across the globe. For example, corruption perception is a sociocultural feature, access to healthcare is a socioeconomic feature, and inflation is an economic feature. These are only a few of the potential predictive features that are analyzed in this work. Forty-five potentially predictive features were initially analyzed for various years and countries of interest. As may seem intuitive, the features that fell under the category of "economic" produced the highest predictive value from the linear regression techniques employed. Over time, however, as more data is collected, it is expected that there will be additional features from both the sociocultural and socioeconomic categories that will begin to stand out as having more predictive impact. Many of these features are nascent in terms of both their development and subsequent collection, as well as the fact that some of these features are not yet collected at a universal level, meaning that the data is missing from some countries or for some years. Iterative data imputation techniques are proposed for accounting for these missing values in ongoing work regarding this dataset for predicting refugee flows.
Recommended Citation
Mead, Esther, "Creating a Machine Learning Model for the Prediction of Refugee Flows" (2021). Theses and Dissertations. 984.
https://research.ualr.edu/etd/984
