Text Mining Freud’s Complete Works: A Lexical-Affective Cartography across Volumes
Abstract
Abstract Sigmund Freud’s work constitutes one of the most influential theoretical corpora in the humanities and health sciences, yet its scope and conceptual density make it difficult to identify large-scale patterns distributed across the texts. This study presents an exploratory text-mining analysis of the Portuguese edition of Freud’s Complete Works, published by Editora Imago, comprising 23 editorial volumes spanning the period from 1886 to 1939. After textual curation, the final corpus retained 12,524,088 characters and 4,145 pages and was analyzed through lexical frequency, n-grams, directed bigram networks, lexical polarity, NRC Emotion Lexicon categories, correlations between psychoanalytic keywords and emotion categories, contextual enrichment analysis, and principal component analysis. The results show that the computational approach recovered recognizable conceptual nuclei of Freudian theory, including dream, sexuality, symptom, neurosis, hysteria, unconscious, desire, pleasure, reality, and childhood. The longitudinal distribution of these terms indicates that key concepts do not appear homogeneously across the collection, but vary across editorial moments, suggesting shifts in lexical emphasis consistent with the historical and theoretical development of psychoanalysis. Lexicon-based affective analyses revealed oscillations in polarity and multidimensional emotional profiles involving positivity, negativity, trust, anticipation, fear, sadness, anger, disgust, joy, and surprise. Psychoanalytic keywords were organized into distinct affective-semantic fields, separating clinical-symptomatic nuclei from dream-interpretive and metapsychological fields. PCA further indicated that different parts of the collection share comparable affective signatures. Taken together, these findings suggest that Freud’s work exhibits a measurable longitudinal lexical-affective organization. This exploratory computational cartography complements traditional theoretical interpretation and opens new possibilities for digital humanities approaches to psychoanalysis.
Fields of Study & Subjects
Mental Health via Writing
Authorship Attribution and Profiling
Sentiment Analysis and Opinion Mining
Publication details
- Journal / source
- Research Square
- Publisher
- Springer Science and Business Media LLC
- Publication date
- 3 July 2026
- Document type
- Preprint