On the Keyword Extraction and Bias Analysis, Graph-based Exploration and Data Augmentation for Abusive Language Detection in Low-Resource Settings

Resumen

PhD thesis in Computer Science focused on Natural Language Processing, written by Gretel Liz De la Peña Sarracén under the supervision of Prof. Paolo Rosso in the Universitat Politècnica de València (Spain). This thesis addresses the abusive language detection specifically emphasizing low-resource settings. The results highlight the potential of graph neural networks models and data augmentation techniques for improving this task. The thesis defense was held virtually on March 6th, 2024. The doctoral committee was composed by Dr. Arkaitz Zubiaga (Queen Mary University of London), Dra. Sara Tonelli (Fondazione Bruno Kessler), and Dra. Aitziber Atutxa (University of the Basque Country). An international mention was achieved, and the work was graded as excellent and awarded Cum Laude.