Maskinlæring som politologisk værktøj
DOI:
https://doi.org/10.7146/politica.v51i2.131144Abstract
Maskinlæring er en metodisk tilgang til databehandling, som vinder indpas i den politologiske forskning og offentlige forvaltning. Her har tilgangen et lovende potentiale til at lave forudsigelser om eksempelvis brugeres og borgeres senere adfærd, hvilket blandt andet kan bruges til målretning af tidlige indsatser. Men hvad er maskinlæring mere konkret, og hvordan anvender man maskinlæring i praksis? I artiklen introducerer vi kernebegreber i relation til maskinlæring. Vi introducerer maskinlæringsalgoritmer i form af klassifikationstræer. Artiklens pointer illustrerer vi undervejs med et konkret eksempel på anvendelse af maskinlæring i dansk offentlig forvaltning, hvor maskinlæring bliver brugt til at forudsige uddannelsesfrafald på Københavns Professionshøjskole. Afslutningsvist diskuterer vi metodiske styrker og svagheder ved maskinlæring i en samfundsvidenskabelig kontekst.
References
Angrist, Joshua D. og Jörn-Steffen Pischke (2015). Mastering ‘Metrics: The Path from Cause to Effect. Princeton University Press.
Athey, Susan og Guido Imbens (2016). The state of applied econometrics-causality and policy evaluation, ArXiv: 1607.00699.
Bach, Alexander og Jesper Svejgaard (2017). Maskinlæring på skolebænken. Speciale, Institut for Statskundskab, Københavns Universitet.
Berk, Richard (2012). Criminal Justice Forecasts of Risk: A Machine Learning Approach. Springer Science & Business Media.
Breiman, Leo (2001). Statistical modelling: The two cultures. Statistical science 16 (3): 199-231.
Chandler, Dana, Steven D. Levitt og John A. List (2011). Predicting and preventing shootings among at-risk youth. The American Economic Review 101 (3): 288–292.
Conway, Drew og John Myles White (2012). Machine Learning for Hackers. Sebastopol: O’Reilly.
Foster, Ian, Rayid Ghani, Ron S. Jarmin, Frauke Kreuter og Julia Lane (2016). Big Data and Social Science. Chapman: Hall/CRC.
Friedman, Jerome H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics 29 (5): 1189-1232.
Gelman, Andrew og Eric Loken (2013). The garden of forking paths: Why multiple comparisons can be a problem, even when there is no “fishing expedition” or “p-hacking” and the research hypothesis was posited ahead of time. Department of Statistics, Columbia University.
Hariri, Jacob Gerner (2012). Kausal inferens i statskundskaben. Politica 44 (2): 184-201.
Hastie, Trevor, Robert Tibshirani og Jerome Friedman (2009). The Elements of Statistical Learning. Springer.
Hill, Daniel W. og Zachary M. Jones (2014). An empirical evaluation of explanations for state repression. American Political Science Review 108 (3): 661-687.
Hofman, Jake M., Amit Sharma og Duncan J. Watts (2017). Prediction and explanation in social systems. Science 355 (6324): 486-488.
James, Gareth, Daniela Witten, Trevor Hastie og Robert Tibshirani (2013). An Introduction to Statistical Learning. Springer.
Kitchin, Rob (2014). Big data, new epistemologies and paradigm shifts. Big Data & Society 1 (1). https://doi.org/10.1177/2053951714528481.
Kleinberg, Jon, Himabindu Lakkaraju, Jure Leskovec, Jens Ludwig og Sendhil Mullainathan (2017). Human Decisions and Machine Predictions.
Kleinberg, Jon, Jens Ludwig, Sendhil Mullainathan og Ziad Obermeyer (2015). Prediction policy problems. American Economic Review 105 (5): 491-495.
Lantz, Brett (2015). Machine Learning with R. Birmingham: Packt Publishing
Montgomery, Jacob M. og Santiago Olivella (2018). Tree‐based models for political science data. American Journal of Political Science 62 (3): 729-744.
Mullainathan, Sendhil og Jann Spiess (2017). Machine learning: An applied econometric approach. Journal of Economic Perspectives 31 (2): 87-106.
Murphy, Kevin P. (2012). Machine Learning: A Probabilistic Perspective. MIT press.
Obermeyer, Ziad og Ezekiel J. Emanuel (2016). Predicting the future: Big data, machine learning, and clinical medicine. The New England Journal of Medicine 375 (13): 1216-1219.
Perry, Walt L., Brian McInnis, Carter C. Price, Susan C. Smith og John S. Hollywood (2013). Predictive Policing: The Role of Crime Forecasting in Law Enforcement Operations. Rand Corporation.
Rosenbaum, Paul R. og Donald B. Rubin (1983). The central role of the propensity score in observational studies for causal effects. Biometrika 70 (1): 41-55.
Samii, Cyrus (2016). Causal empiricism in quantitative research. The Journal of Politics 78 (3): 941-955.
Samuel, Arthur L. (1959). Some studies in machine learning using the game of checkers. IBM Journal of Research and Development 3 (3): 210-229.
Shalev-Shwartz, Shai og Shai Ben-David (2014). Understanding Machine Learning: From Theory to Algorithms. New York: Cambridge University Press.
Varian, Hal R. (2014). Big data: New tricks for econometrics. The Journal of Economic Perspectives 28 (2): 3-27.
Wooldridge, Jeffrey M. (2009). Introductory Econometrics: A Modern Approach. South-Western Cengage Learning.
