Back close

An Approach Towards Abstractive Summary Generation and Knowledge Graph Representation

Publication Type : Conference Paper

Publisher : Springer Nature Singapore

Source : Lecture Notes in Electrical Engineering

Url : https://doi.org/10.1007/978-981-97-3090-2_19

Campus : Amaravati

School : School of Computing

Year : 2025

Abstract : With the proliferation of blogs, news stories, and reports, the extraction of useful information from this extensive collection of textual documents is a tedious task. Automatic text summarization provides an effective solution for condensing these documents into concise summaries while preserving essential information and meaning. Numerous noteworthy summarization models have been proposed to address various challenges, including saliency, fluency, human readability, and the generation of high-quality summaries. In this study, we introduce the Text-To-Text Transfer Transformer (T5) model for the task of abstractive summarization with knowledge representation. The experimental results demonstrate that the T5 model produces summaries that are more conceptual, comprehensible, and abstractive. To assess the quality of the generated summaries, we consider the ROUGE and BLEU scores.

Cite this Research Publication : Prottay Kumar Adhikary, Pankaj Dadure, Riyanka Manna, Partha Pakray, An Approach Towards Abstractive Summary Generation and Knowledge Graph Representation, Lecture Notes in Electrical Engineering, Springer Nature Singapore, 2025, https://doi.org/10.1007/978-981-97-3090-2_19

Admissions Apply Now