ariboyarsky / ml-political-methodology

This repo holds code for a paper which discusses applications of Machine Learning in Political Methodology.

ML-Political-Methodology

This repo holds code for a paper which discusses applications of Machine Learning in Political Methodology.

Please note that this code base is in the form of an R Studio Project. Feel free to use this code as you may, though please cite the author as such: Boyarsky, Ariel. "ML-Political-Methodology." Github Repository. 2016.

Abstract

we look at how machine learning techniques can aid researchers through text analysis and casual identification. To accomplish this we look at how sentiment analysis may be employed to study political polarization, and how new techniques in data analysis may solve questions of causal inference. We also explore the literature and theory behind these methodologies using research in political science, computer science, and statistics. We hope that this paper serves as an introduction to the use of advanced computational methods in political analysis.

This paper was written for a Big Data course at the George Washington University. If you would like a copy, please email the author at [email protected].

Coding References

Feinerer, Ingo, and Hornik, Kurt. "Introduction to the RKEA Package." Cran.R. 2015. https://cran.r-project.org/web/packages/RKEA/vignettes/kea.pdf

Ho, Daniel E. "MatchIt: nonparametric preprocessing for parametric causal inference." PhD diss., Departments of Mental Health and Biostatistics, Johns Hopkins Bloomberg School of Public Health, 1737.

Minqing Hu and Bing Liu. "Mining and Summarizing Customer Reviews." Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-2004), Aug 22-25, 2004, Seattle, Washington, USA