The AraPoemBERT model is introduced as a pretrained language model exclusively for Arabic poetry text. It outperforms other Arabic language models in tasks related to Arabic poetry analysis. The complexity of Arabic poetry requires advanced computational models for accurate analysis. The dataset used contains over 2.09 million verses with attributes like meter, sub-meter, poet, rhyme, and topic. AraPoemBERT achieves state-of-the-art results in gender classification, sentiment analysis, and meter classification tasks.
Til et andet sprog
fra kildeindhold
arxiv.org
Vigtigste indsigter udtrukket fra
by Faisal Qarah kl. arxiv.org 03-20-2024
https://arxiv.org/pdf/2403.12392.pdfDybere Forespørgsler