CoToHiLi: computational tools for historical linguistics

• Alina Maria Cristea ^[1] ; Anca Dinu ^[1] ; Liviu P. Dinu ^[1] ; Simona Georgescu ^[1] ; Ana Sabina Uban ^[1] ; Laurentiu Zoicas ^[1]
  1. [1] University of Bucharest

     University of Bucharest

     Sector 3, Rumanía

     
• Localización: SEPLN-PD 2022: Annual Conference of the Spanish Association for Natural Language Processing 2022: Projects and Demonstrations / Miguel Á. Alonso (ed. lit.), Margarita Alonso-Ramos (ed. lit.), Carlos Gómez Rodríguez (ed. lit.), David Vilares Calvo (ed. lit.), Jesús Vilares (ed. lit.), 2022, págs. 31-34
• Idioma: inglés
• Títulos paralelos:
  • CoToHiLi: herramientas computacionales para la lingüística histórica
• Enlaces
  • Texto completo (pdf)
• Resumen

  • This project represents a computational framework for historical linguistics. The general purpose of theCoToHiLi project is to integrate expert knowledge and computational power to address cognate identification,cognate-borrowing discrimination, Latin protoword reconstruction and semantic divergence. The goal of theproject is twofold: 1) to automate certain parts of the traditional work-flow of the comparative method (suchas the collection of data or the automatic alignment based on predefined or inferred rules), and 2) to bringnew insights or avenues of investigation, which might not be easily accessible otherwise (e.g., the automaticidentification of patterns and regularities in large amounts of data). The project will provide tools for the mainRomance kernel group (French, Italian, Portuguese, Romanian, Spanish), as well as Latin. The methodologiesand computational tools proposed could also serve as a basis for further development for other comparablelanguage families, including less studied languages, with scarce resources available.