Domain Specific Text Summarization for Hindi Using Transformer-Based Models

The rapid increase of digital information in India has generated an increasing demand for effective techniques to synthesize domain-specific literature in natural languages. In this research, the challenges and advancements in domain-specific text synthesis and summarization for Indian languages are explored. It summarizes deep learning along-with machine learning methods and focuses especially on customizing techniques to handle the linguistic complexity. The research highlights issues related to the lack of data available with low-resource languages. Furthermore, the study investigates the latest developments done in transformerbased models for text synthesis, including GPT-3 and BERT. The performance of the proposed model is examined in terms of different parameters such as precision (70.59%), recall (58.81%), and F1 score (62.34%}. Besides addressing these issues, it promotes digital communication and information sharing that is inclusive in India.