International Journal of Science, Engineering and TechnologyAuthors: Dr. Brij Mohan, Sagar Chaudhary
Abstract: Massive amounts of textual content every day are produced due to the rapid expansion of digital news channels. Quickly drawing significant conclusions from lengthy news articles may be a challenge for readers, journalists, and academics. The problem of condensing long articles into brief and instructive summaries has been solved by automating text summarizing, which preserves the fundamental meaning and context. The design and implementation of an AI chatbot-based system for automatic text summarization of news articles are presented in this research paper using deep learning and NLP approaches. The suggested system provides support for abstract, hybrid, and extractive summarization techniques. Despite the sophistication of transformer-based models like BERT[4, T5[7], and PEGASUS[7] Abstract summarizing is done by using traditional extractive techniques like TF- to provide summaries that are human-like. The statistical significance of IDF[2] and TextRank[8] is used to identify significant phrases. The modular system architecture includes preprocessing, summary engine, evaluation module, and interactive chatbot interface, making it easy for users to enter news items and receive summarized results instantly.
DOI: http://doi.org/10.5281/zenodo.21418085
