The 2024 Conference on Empirical Methods in Natural Language Processing

At EMNLP 2024, Sania Aftarha presented RoBERT2VecTM: A Novel Approach for Topic Extraction in Islamic Studies, co-authored by Luca Gagliardelli, Amina El Ganadi, Federico Ruozzi, and Sonia Bergamaschi (University of Modena and Reggio Emilia)

The study introduces RoBERT2VecTM, combining RoBERTa and Doc2Vec to enhance topic modeling in Hadith studies. The approach improves semantic analysis, generating more coherent and diverse Arabic topics. The research also highlights the role of lemmatization and stopwords in refining results.