COMPEL: Computational Analysis of Peripheral Literatures
Recent technologies like Large Language Models have brought major progress in Natural Language Processing. However, some scenarios present particular challenges, like historical language varieties. Moreover, some tasks like the automatic analysis of literary texts also present specific challenges and are less commonly studied. Indeed, the analysis of literary data across various languages is essential for understanding cultural practices and products throughout history. However, many languages lack adequate computational resources towards this end. The COMPEL project is addressing literature in Galician, a language included in the European Charter for Regional or Minority Languages, to enhance computational literary analysis resources in lesser-studied traditions.
Objectives
The project goal is to increase the resources enabling the computational analysis of literary texts in lesser-studied traditions, focusing on Galician, a rich but peripheral tradition at the crossroads of two major romance literatures, Portuguese and Spanish. The project will focus on two genres, poetry and narrative, and two main Galician literature movements in the 19th century and at the turn of the 20th century; public domain texts will allow us to release project resources under open licenses. We will develop methods to compare texts from several literary traditions cross-lingually. The project aims to promote inclusive literary history while contributing to European linguistic diversity goals. It advocates for European identity by facilitating smarter access to cultural heritage, aligning with current EU priorities.
Project
/research/projects/computational-analysis-of-peripheral-literatures
<p>Recent technologies like Large Language Models have brought major progress in Natural Language Processing. However, some scenarios present particular challenges, like historical language varieties. Moreover, some tasks like the automatic analysis of literary texts also present specific challenges and are less commonly studied. Indeed, the analysis of literary data across various languages is essential for understanding cultural practices and products throughout history. However, many languages lack adequate computational resources towards this end. The COMPEL project is addressing literature in Galician, a language included in the European Charter for Regional or Minority Languages, to enhance computational literary analysis resources in lesser-studied traditions.</p><p>The project goal is to increase the resources enabling the computational analysis of literary texts in lesser-studied traditions, focusing on Galician, a rich but peripheral tradition at the crossroads of two major romance literatures, Portuguese and Spanish. The project will focus on two genres, poetry and narrative, and two main Galician literature movements in the 19th century and at the turn of the 20th century; public domain texts will allow us to release project resources under open licenses. We will develop methods to compare texts from several literary traditions cross-lingually. The project aims to promote inclusive literary history while contributing to European linguistic diversity goals. It advocates for European identity by facilitating smarter access to cultural heritage, aligning with current EU priorities.</p> - 101149659 - Pablo Ruiz Fabo - Pablo Gamallo Otero
projects_en