Topics (Automated Content Analysis)

Topics describe the main issue discussed in an article, for example: Does article deal with politics, economics or sports? Field of application/theoretical foundation: In context “Agenda Setting”, studies analyze which issues are on public agenda. “News Values”, may why some topics more prominently covered than others. References/combination other methods data collection: Many combine manual inspection their automated detection. Quinn et al. (2010) demonstrate analyses legislative speeches how increase validity results. Similarly, Hase (2020) use content analysis to find and map similar coding is then conducted. Such combinations contribute a better detailed understanding by themselves. The datasets referred table described following paragraph: Puschmann (2019a) uses New York Times articles (1996-2006, N = 30,862) as well from Die Zeit (2011-2016, 377) identify using supervised machine learning. another tutorial, (2019b) Sherlock Holmes stories (18th century, 12), debate transcripts (1970-2017, 7,897) apply LDA structural topic modeling. her tutorials, Silge (2018a, 2018b) also 12) news corpus containing comments (2006-ongoing, 100,000). Robinson modeling Associated Press (1992, 2,246) books Dickens, Wells, Verne Austen 4). Roberts (2019) blogposts (2008, 13,248) Watanabe Müller newspaper Guardian (2016, 6,000). Van Atteveldt Welbers (2019, 2020) State Union (1981-2017, 10 1789-2017, 58) analyses. Lastly, Wiedemann Niekler (2017) same (1790-2017, 223). Table 1. Measurement “Topics” analysis. Author(s) Sample Procedure Formal check benchmark* Code (a) Newspaper (b) Supervised learning Reported http://inhaltsanalyse-mit-r.de/maschinelles_lernen.html (c) United Nations General Debate Transcripts modeling; Not reported http://inhaltsanalyse-mit-r.de/themenmodelle.html (2018a) & (2018b) News t Structural https://juliasilge.com/blog/sherlock-holmes-stm/ https://juliasilge.com/blog/evaluating-stm/ Books https://www.tidytextmining.com/topicmodeling.html Blogposts https://www.jstatsoft.org/article/view/v091i02 https://tutorials.quanteda.io/machine-learning/topicmodel/ van https://github.com/ccs-amsterdam/r-course-material/blob/master/tutorials/r_text_stm.md https://github.com/ccs-amsterdam/r-course-material/blob/master/tutorials/r_text_lda.md https://tm4ss.github.io/docs/Tutorial_6_Topic_Models.html https://tm4ss.github.io/docs/Tutorial_7_Klassifikation.html *Please note that many sources listed here tutorials conducted – therefore not focused validation Readers should simply read this column indication terms they can refer if interested References Hase, V., Engelke, K., Kieslich, K. (2020). things we fear. Combining uncover themes, threats fear-related news. Journalism Studies, 21(10), 1384-1402. Puschmann, C. (2019). Automatisierte Inhaltsanalyse mit R. Retrieved http://inhaltsanalyse-mit-r.de/index.html Quinn, M., Monroe, B. L., Colaresi, Crespin, M. H., Radev, D. (2010). How political attention minimal assumptions costs. American Journal Political Science, 54(1), 209–228. Roberts, E., Stewart, Tingley, stm: An R Package Topic Model. Statistical Software, 91(2), 1–40. Silge, J. (2018a). game afoot! stories. (2018b). Training, evaluating, interpreting models. J., Robinson, Text Mining A tidy approach. https://www.tidytextmining.com/ Atteveldt, W., Welbers, Modeling. Fitting models Watanabe, Müller, S. Quanteda tutorials. https://tutorials.quanteda.io/ Wiedemann, G., Niekler, A. (2017). Hands-on: five day text mining course humanists social scientists Proceedings 1st Workshop Teaching NLP Digital Humanities (Teach4DH@GSCL 2017), Berlin. https://tm4ss.github.io/docs/index.html