Sem-Rouge: Graph-Based Embedding for Automated Text Summarization with Using Large Language Models.
Saved in:
| Title: | Sem-Rouge: Graph-Based Embedding for Automated Text Summarization with Using Large Language Models. |
|---|---|
| Authors: | Pulari, Sini Raj1 (AUTHOR), Umadevi, Maramreddy1 (AUTHOR), Vasudevan, Shriram K2 (AUTHOR) shriramkv@gmail.com |
| Source: | Journal of Intelligent & Fuzzy Systems. Oct2025, Vol. 49 Issue 4, p1057-1070. 14p. |
| Subjects: | Automatic summarization, Artificial intelligence, Language models, Text summarization, Generative pre-trained transformers, Natural language processing, Electronic newspapers |
| Abstract: | Information accessibility has been transformed in the field of Artificial Intelligence (AI), particularly in natural language processing (NLP), owing to the widespread use of technologies such as ChatGPT. This paper explores the field of human-directed AI solutions with particular emphasis on newspaper summarization, a useful tool in today's busy world. Utilizing comprehensive models (LLMs), we explore extractive and abstractive summarization methods. To maximize the LLM performance, our strategy entails creating a customized news dataset enhanced with human-centric summaries and using cutting-edge data preparation techniques. To improve the accuracy assessment, we present a modified evaluation metric called Sem-rouge, which augments established units of measurement. In a comparative analysis, it was noticed that the proposed metric can highlight both syntactic and semantic similarities; hence, the metric is suitable for both extractive and abstractive summarization methods. We highlight the significance of dataset selection, data processing methods, and assessment criteria in fine-tuning auto-generated summaries using rigorous comparison analysis. Further studies will focus on improving semantic similarity techniques, integrating advanced models such as The BERT algorithm or Generative Pre-trained Transformer algorithm, and overcoming challenges such as overfitting. Finally, our study emphasizes the importance of meticulously training models and modifying them frequently to enhance automated summarization skills. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Intelligent & Fuzzy Systems is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Information accessibility has been transformed in the field of Artificial Intelligence (AI), particularly in natural language processing (NLP), owing to the widespread use of technologies such as ChatGPT. This paper explores the field of human-directed AI solutions with particular emphasis on newspaper summarization, a useful tool in today's busy world. Utilizing comprehensive models (LLMs), we explore extractive and abstractive summarization methods. To maximize the LLM performance, our strategy entails creating a customized news dataset enhanced with human-centric summaries and using cutting-edge data preparation techniques. To improve the accuracy assessment, we present a modified evaluation metric called Sem-rouge, which augments established units of measurement. In a comparative analysis, it was noticed that the proposed metric can highlight both syntactic and semantic similarities; hence, the metric is suitable for both extractive and abstractive summarization methods. We highlight the significance of dataset selection, data processing methods, and assessment criteria in fine-tuning auto-generated summaries using rigorous comparison analysis. Further studies will focus on improving semantic similarity techniques, integrating advanced models such as The BERT algorithm or Generative Pre-trained Transformer algorithm, and overcoming challenges such as overfitting. Finally, our study emphasizes the importance of meticulously training models and modifying them frequently to enhance automated summarization skills. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 10641246 |
| DOI: | 10.1177/18758967251353031 |
