The field of computational social science is rapidly growing, due in part to the availability of data analysis methods such as natural language processing. Many CSS researchers are familiar with the basics of NLP such as word frequency analysis, but more
advanced methods such as event extraction remain difficult for beginners to understand and use correctly. Furthermore, new NLP methods are often shared in the form of academic papers and code repositories without a structured explanation of how researchers can apply the methods to their own work. We propose a tutorial series that will provide an introduction to more advanced NLP methods with a broad appeal to early CSS researchers, including word embeddings, event extraction, large-scale language models, and causal inference. We will invite current experts in NLP to present their newly developed methods in an interactive format, which will allow participants to explore the utility of the methods as applied to social science research. The interactive format of the tutorials will also allow researchers to connect via shared interests in methods, which can lead to sharing of ideas and collaborations. The materials for the tutorials will be made publicly available for later sharing in the broader CSS community.
PhD Student, College of Information and Computer Sciences, University of Massachusetts - Amherst
Postdoctoral Research Fellow, University of Michigan