Towards an Author-Topic-Term-Model Visualization of 100 Years of German Sociological Society Proceedings

摘要

Author co-citation studies employ factor analysis to reduce high-dimensionalco-citation matrices to low-dimensional and possibly interpretable factors, butthese studies do not use any information from the text bodies of publications.We hypothesise that term frequencies may yield useful information forscientometric analysis. In our work we ask if word features in combination withBayesian analysis allow well-founded science mapping studies. This work goesback to the roots of Mosteller and Wallace's (1964) statistical text analysisusing word frequency features and a Bayesian inference approach, tough withdifferent goals. To answer our research question we (i) introduce a new dataset on which the experiments are carried out, (ii) describe the Bayesian modelemployed for inference and (iii) present first results of the analysis.