Topic modeling (LDA) applied to an environmental controversy: Isotopies of environmental discourse about fire in the Paraná River islands, Argentina

This article explores the application of Latent Dirichlet Allocation (LDA), a quantitative topic modeling technique, to the environmental controversy surrounding the fires on the Paraná River islands from 2020 to 2022. It examines how integrating qualitative and quantitative methods can enhance rese...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Preiti, Francisco José, Alomar, Francisco Javier
Formato: Artículo revista
Lenguaje:Español
Publicado: Facultad de Ciencia Política y Relaciones Internacionales, Universidad Nacional de Rosario 2024
Materias:
LDA
Acceso en línea:https://latrama.unr.edu.ar/index.php/trama/article/view/846
Aporte de:
Descripción
Sumario:This article explores the application of Latent Dirichlet Allocation (LDA), a quantitative topic modeling technique, to the environmental controversy surrounding the fires on the Paraná River islands from 2020 to 2022. It examines how integrating qualitative and quantitative methods can enhance research in anthropology, political ecology, and discourse analysis. The study utilizes LDA to investigate the formation of isotopies—emergent semantic clusters from interviews and an assembly held during environmental protests in the City of Rosario, Argentina. Findings highlight key metrics, topics, and insights crucial for defining isotopies, which are connected to the organizational and political identities of the groups involved, the dichotomy between conservation and development, and the interplay between global and local discourses. This case study thereby characterizes the environmental discourse shaped by participants in the conflict dynamics. The paper advocates for this methodological integration to deepen the analysis of environmental controversies