Skip to content

Exam project

Jump to bottom
Andreas Bjerre-Nielsen edited this page Dec 17, 2018 · 1 revision

Project inspiration for Topics in Social Data Science

Relevant research (will be updated)

For getting some inspiration we recommend that your read a few recent research articles that have used machine learning for conducting social science. Below is a selection of topics and associated articles.

Blumenstock, J., Cadamuro, G., and On, R. (2015). Predicting poverty and wealth from mobile phone metadata. Science, 350(6264):1073--1076

Stephens-Davidowitz, S. (2013b). Unreported Victims of an Economic Downturn

Glaeser, E. L., Kominers, S. D., Luca, M., and Naik, N. (2015). Big Data and Big Cities: The Promises and Limitations of Improved Measures of Urban Life. Working Paper 21778, National Bureau of Economic Research

Gebru, T., Krause, J., Wang, Y., Chen, D., Deng, J., Aiden, E. L., and Fei-Fei, L. (2017). Using deep learning and Google Street View to estimate the demographic makeup of neighborhoods across the United States.

Wu, A. (2017). Gender Stereotyping in Academia: Evidence from Economics Job Market Rumors Forum. SSRN Scholarly Paper ID 3051462, Social Science Research Network, Rochester, NY

Gentzkow, M., Shapiro, J. M., and Taddy, M. (2016). Measuring Polarization in High-Dimensional Data: Method and Application to Congressional Speech. Working Paper 22423, National Bureau of Economic Research

Hansen, S., McMahon, M. and Prat, A., 2017. Transparency and deliberation within the FOMC: a computational linguistics approach. The Quarterly Journal of Economics, 133(2), pp.801-870.

Mønsted, B., Sapieżyński, P., Ferrara, E. and Lehmann, S., 2017. Evidence of complex contagion of information in social media: An experiment using Twitter bots. PloS one, 12(9), p.e0184148.

Concrete ideas and data (will be updated)

Grading

The grade for this course is exclusively determined by the project handed in. The project will be judged on a number of dimensions, these include:

  • how the data was obtained (setting up new data collection);
  • how the tools for working with networks, geo-data, text and machine learning are applied (at least one must be used, two is recommended);
  • how the methods are applied and which methods are used;
  • how results are explained (writing, figures, tables with model output etc.);
  • the research question and its originality and how it is answered.

Requirements for project

The exam projects have a number of requirements that must be met, these are:

  • Research question (hand in April 30, max half page)
  • Groups with up to four members, you choose
  • Project formalia
    • Report (.pdf file)
      • The style should be like a light written research article (brief literature review, references to methods etc.)
      • Grading will be based on this report but process should be document in Jupyter Notebook.
      • The following maximum number of pages (normalsider): 1 pax: 12 pages, 2 pax: 16 pages, 3 pax: 20 pages, 4 pax: 24 pages.
    • Documentation in Jupyter Notebook (.ipynb file)
  • Some advice
    • It is more important that you spent time on calibrating and validating the models you work with rather than using as many models as possible.

Clone this wiki locally