Features in extractive supervised single-document summarization: case of Persian news

Rezaei, Hosein; Mirhosseini, Seyed Amid Moeinzadeh; Shahgholian, Azar; Saraee, Mohamad

Abstract

Text summarization has been one of the most challenging areas of research in NLP. Much effort has been made to overcome this challenge by using either abstractive or extractive methods. Extractive methods are preferable due to their simplicity compared with the more elaborate abstractive methods. In extractive supervised single-document approaches, the system will not generate sentences. Instead, via supervised learning, it learns how to score sentences within the document based on some textual features and subsequently selects those with the highest rank. Therefore, the core objective is ranking, which enormously depends on the document structure and context. These dependencies have been unnoticed by many state-of-the-art solutions. In this work, document-related features such as topic and relative length are integrated into the vectors of every sentence to enhance the quality of summaries. Our experiment results show that the system takes contextual and structural patterns into account, which will increase the precision of the learned model. Consequently, our method will produce more comprehensive and concise summaries.

Citation

Rezaei, H., Mirhosseini, S. A. M., Shahgholian, A., & Saraee, M. (in press). Features in extractive supervised single-document summarization: case of Persian news. Language Resources and Evaluation, 1 - 19. https://doi.org/10.1007/s10579-024-09739-7

• Journal Article Type: Article
• Acceptance Date: Apr 5, 2024
• Online Publication Date: May 8, 2024
• Deposit Date: May 22, 2024
• Publicly Available Date: May 28, 2024
• Journal: Language Resources and Evaluation
• Print ISSN: 1574-020X
• Electronic ISSN: 1574-0218
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• Pages: 1 - 19
• DOI: https://doi.org/10.1007/s10579-024-09739-7
• Keywords: Supervised extractive summarization · Machine learning · Regression · Feature extraction · Natural language processing
• Publisher URL: https://link.springer.com/article/10.1007/s10579-024-09739-7