Extraction of Key Concept Relevance Graphs from Fourteen Decades of Psychoanalytic Journal Publications

[abstract]

Automatically tracing the history of concepts and terms by text mining diachronic corpora is a new multidisciplinary area of research. An essential step in the pipeline of such research is calculating the relevance of terms, with most work relying on simple frequency measures. In this study, we extract relevance graphs of key psychoanalytic concepts from a corpus built for this task, one that represents the entire history of psychoanalysis. Different measures of relevance are examined and shown to tell a different story; when combined, however, they provide a clearer picture of a term's significance over time.