BeliN: A Novel Corpus for Bengali Religious News Headline Generation using Contextual Feature Fusion

2025-01-02

AI Summary

This work addresses the low-resource, multi-dimensional modeling challenge of Bangla religious news headline generation. We propose MultiGen, the first model to jointly encode article content with multi-faceted contextual features—including news category, sentiment polarity, and core topic—to enable fine-grained semantic guidance. To support this research, we introduce and publicly release BeliN, the first benchmark corpus for Bangla religious news headline generation, featuring human-annotated headlines, categories, sentiment labels, and thematic tags. MultiGen builds upon multilingual pretrained language models (e.g., BanglaT5, mBART, mT5, mT0) and incorporates a lightweight multi-source feature fusion architecture. On BeliN, MultiGen significantly outperforms content-only baselines: BLEU-4 improves by +2.53 to 18.61, and ROUGE-L by +1.11 to 24.19. Both code and dataset are openly available.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Multimodal Learning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Automatic text summarization, particularly headline generation, remains a critical yet underexplored area for Bengali religious news. Existing approaches to headline generation typically rely solely on the article content, overlooking crucial contextual features such as sentiment, category, and aspect. This limitation significantly hinders their effectiveness and overall performance. This study addresses this limitation by introducing a novel corpus, BeliN (Bengali Religious News) - comprising religious news articles from prominent Bangladeshi online newspapers, and MultiGen - a contextual multi-input feature fusion headline generation approach. Leveraging transformer-based pre-trained language models such as BanglaT5, mBART, mT5, and mT0, MultiGen integrates additional contextual features - including category, aspect, and sentiment - with the news content. This fusion enables the model to capture critical contextual information often overlooked by traditional methods. Experimental results demonstrate the superiority of MultiGen over the baseline approach that uses only news content, achieving a BLEU score of 18.61 and ROUGE-L score of 24.19, compared to baseline approach scores of 16.08 and 23.08, respectively. These findings underscore the importance of incorporating contextual features in headline generation for low-resource languages. By bridging linguistic and cultural gaps, this research advances natural language processing for Bengali and other underrepresented languages. To promote reproducibility and further exploration, the dataset and implementation code are publicly accessible at https://github.com/akabircs/BeliN.

Problem

• Automatic Title Generation
• Bangla Religious News
• Multi-faceted Information Incorporation

Innovation

• BeliN Database
• MultiGen Method
• Contextual Title Generation