Abstract

MANRIQUE-GOMEZ, Laura; MONTES, Tony  and  MANRIQUE, Rubén. Gender Semantics and Historical Feminisms: An Interdisciplinary Approach through Natural Language Processing. rev.estud.soc. [online]. 2025, n.93, pp.19-38.  Epub Aug 19, 2025. ISSN 0123-885X.  https://doi.org/10.7440/res93.2025.02.

This article explores the evolution of gender semantics in 19th-century Latin America, focusing on the semantic nuances of the word women. The study’s primary aim is to present the results of an interdisciplinary methodology that reveals historical gender concepts embedded in language, analyzed through the dual lens of social sciences and artificial intelligence. This study uses Machine Learning techniques to analyze historical texts, specifically employing Natural Language Processing to detect semantic shifts, part-of-speech tagging, and named entity recognition to identify key gender vocabulary. Additionally, an n-gram approach was employed to recognize the most frequent terms associated with target words. The methodology was applied to a self-collected historical corpus from Latin American Spanish newspapers, demonstrating the effectiveness of these technologies in processing extensive collections of written sources. The article reveals how artificial intelligence tools can elucidate underlying gender ideas in historical written texts, offering empirical insights into the historical inequalities in linguistic representation. By comparing the newspaper dataset results with a specific literary work, the Colombian novel Manuela by Eugenio Díaz Castro (1859), the research highlights latent feminist tensions and revolutionary ideas contesting societal norms. The article does not provide a definitive historical or literary analysis. Instead, it invites social scientists to engage in the next phase of research by illustrating the potential of artificial intelligence in enhancing interpretability and critical analysis of historical narratives through interdisciplinary collaboration. This article contributes to historical feminist debates by presenting an original framework synthesizing qualitative and computational methods, open datasets, and code, thus expanding the possibilities of traditional historiography. It concludes with reflections on improving gender semantic studies, emphasizing how this integration of disciplines can propel future research directions in critical cultural inquiries.

Keywords : artificial intelligence; gender semantics; historical feminisms; machine learning; natural language processing; 19th-century Latin America.

        · abstract in Spanish | Portuguese     · text in English     · English ( pdf )