Doctoral Position (M/F) Automated Detection of Intertextuality in Literary Corpora

Descrizione

Doctoral Position (M/F) Automated Detection of Intertextuality in Literary Corpora Sintesi in italiano (traduzione automatica): La posizione di dottorato si svolgerà presso il Laboratoire d'informatique Gaspard-Monge (LIGM) all'interno del team ADA, sotto la supervisione del professor Philippe Gambette e con la co-supervisione della professoressa Delphine Amstutz. Il progetto di ricerca si concentra sulla rilevazione automatica dell'intertestualità in grandi corpora letterari, richiedendo competenze algoritmiche e capacità di codifica. È prevista una collaborazione interdisciplinare con esperti di letteratura francese del XVII secolo. I candidati devono avere interesse per l'analisi di testi letterari e competenze in tecniche algoritmiche per elaborare rapidamente centinaia di libri. È richiesta una laurea in informatica o un campo correlato, con un focus su stringologia e analisi testuale. The doctoral research will take place at the Laboratoire d'informatique Gaspard-Monge (LIGM), within the ADA (Discrete Algorithms and Applications) team, under the supervision of Philippe Gambette, professor of computer science, with co-supervision by Delphine Amstutz, an associate professor in 17th-century French literature at Sorbonne University (Cellf). Collaborations are planned with Gregory Kucherov (LIGM) and Romain Jalabert (Cellf). An interest in interdisciplinary collaboration on French literary texts is therefore expected, in addition to algorithmic skills and the ability to code a prototype implementing the developed algorithms. This doctoral project aims to automatically detect intertextuality (quotations, allusions, rewritings, and adaptations) in large literary corpora. The goal is to process, quickly and efficiently, several hundred books using text-based algorithmic techniques (stringology) to search for excerpts from a first text that have been included—with varying degrees of modification—in a second text, while determining whether the identified textual similarities are significant or coincidental. The variability of the texts—due, for example, to 17th-century spelling or digitization errors—and the specific case of texts in verse must be taken into account, as must certain constraints useful for limiting the search space (for example, restricting the search to sources predating the text under study). An evaluation of the relevance of the developed approach and tool will be conducted on a literary corpus in terms of computation time, quality of results, and usability. Two application corpora will be used: one consisting of 17th-century political texts, and the other comprising the poetic works of Marceline Desbordes-Valmore. Annuncio in inglese. Fonte: Euraxess (Commissione europea).