Semantic analysis of web archive historical data the 1983 "Marche pour l'égalité et contre le racisme"
Year: 2024
Type: dissertation
Abstract: Web archives represent an invaluable source for historians and re- searchers in digital humanities to study the relation between a society and its past. However, the surplus of digitalised resources makes it difficult to deal with these sources in terms of navigation, exploration and analysis. To effectively address this matter, it is crucial to adopt an interdisciplinary approach that integrates the ex- pertise and perspectives of both historians and computer scientists. After a careful review of related work, it becomes clear that the assistance provided by Natural Language Processing (NLP) and network analysis tools is significant in this particular context. In this thesis, we use the data obtained from the web archive of INA, which includes webpages referencing the 1983 “ Marche pour l’egalité et contre le racisme ”. This data serves as the foundation for implementing a framework that combines Named Entity Recognition (NER), Topic Modeling, and network analysis. The objective is to provide historians with a navigational tool that facilitates their exploration of extensive corpora. Promising results have been achieved, paving the way for future enhancements to the framework.
Authors: Rendina, Davide, Gebeil, Sophie, Génois, Mathieu, Bellot, Patrice
Language: English
Subfield: Literature and Literary Theory
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: green
Funder: Agence Nationale de la Recherche
OpenAlex ID: W6949053367
Cites:
Cited by:
Related to: