PhD Defense: 'Hyperpartisan News Detection'

Hyperpartisan news detection has been widely studied, yet existing research remains largely English-centric, relies heavily on binary classifications, and often overlooks the pragmatic and linguistic mechanisms through which political bias is constructed.

This PhD thesis addresses these limitations through a comprehensive research approach encompassing systematic mapping, methodological benchmarking, and the development of multilingual resources for Spanish, Italian, and Portuguese. The research also explores methods for neutralizing hyperpartisan content while preserving its underlying ideological stance, contributing to a more nuanced understanding of political bias and its linguistic expression.

Supervisors: Pablo Gamallo Otero and Gaël Dias