Critical Georeferencing of the First Five Novels of Mario Vargas Llosa Using Artificial Intelligence

Mario Vargas Llosa (1936–2025) is the most important contemporary Peruvian writer. His work circulates worldwide, and its exceptional quality was recognized with the Nobel Prize in Literature in 2010. His literary production comprises 20 novels, 3 short story collections, 17 essay collections, and 2 memoirs, as well as a substantial number of journalistic articles. In his novels, space functions as a crucial element of the plot: without it, the unfolding of events would be impossible. This is the case, for instance, with the oppressive atmosphere of the Leoncio Prado Military Academy, which gives rise within its walls to the drama of the interns (cadets); similarly, the bar La Catedral serves as the setting for the conversation between Santiago Zavala and the Ambrosio (the former driver of his father), which constitutes the narrative thread of the novel. Thus, from the Amazonian city of Iquitos to Diego Ferré and Ocharán streets in the Lima district of Miraflores, the places depicted in these novels are not mere backdrops but generative spaces that enable and shape the events narrated.

 

In light of the above, the project “Critical Georeferencing of the First Five Novels of Mario Vargas Llosa Using Artificial Intelligence” seeks to critically georeference the Peruvian locations present in the first five novels of the author: La ciudad y los perros (1963), La casa verde (1966), Conversación en La Catedral (1969), Pantaleón y las visitadoras (1973), and La tía Julia y el escribidor (1977). With the support of Artificial Intelligence (AI) and other technological tools, the project aims to develop a digital platform hosting an interactive map of Peru on which the places mentioned in these texts by the Nobel Prize–winning author will be located. The process will begin by providing the AI with the novels in PDF format and training it to identify the locations referenced in the texts through Toponymic Named Entity Recognition (NER). This will be followed by disambiguation, geocoding, coordinate validation, and, finally, the generation of the interactive map using tools such as Folium or Leaflet.

 

It is important to note that each georeferenced location will be accompanied by an informational pop-up window activated by user interaction. This window will include one or more textual excerpts in which the relevant place is mentioned, as well as an explanation of the significance of the selected space—this constitutes the critical dimension of the georeferencing exercise. Whenever possible, historical photographs from the period in which the novels are set will be displayed alongside contemporary images of the same locations (for example, archival records of the Leoncio Prado Military Academy can be contrasted with current photographs).

 

The primary objective of the project is to consolidate a pedagogical tool within the framework of the Digital Humanities that offers readers of these five novels by Vargas Llosa a dynamic reading experience, fostering greater engagement and deeper comprehension.

 

The paper will present in detail the nature of the project, which is currently developing, its objectives, the motivations behind its implementation, its significance for the development of Digital Humanities in relation to Peruvian Literary Studies, and the methodological procedures to be employed.