Advanced News Archiving System with Machine Learning-Driven Web Scraping and AI-Powered Summarization Using T5, Pegasus, BERT and BART Architectures
Year: 2024
Type: article
Abstract: Data plays a crucial role in the contemporary era of technology, as it is a vital element in the publication of news on the internet or a website. Nevertheless, understanding long reports in order to fully comprehend events can be a challenging endeavor, frequently leading to subjective judgments. The application's architecture integrates the categorization of news stories by day, resulting in a well-organized and readily accessible archive. The application employs the web scraping method, which entails pulling pertinent news articles from numerous internet sources. The application employed sophisticated summarizing libraries, including the BERT, BART, T5 model and Google Pegasus, to condense the information into a succinct and comprehensible style. The T5 model performs exceptionally well in text summarization and other natural language processing tasks because of its text-to-text structure; it is also a very customizable language model. Google Pegasus, an expert in abstractive summarizing, uses self-attention mechanisms and rigorous pre-training to generate high-quality, concise news summaries. To summarize, these are the most important parts of our app's process. When it comes to collecting, storing, and summarizing news articles, the system has you covered. In addition, it will offer a straightforward design that makes it simple to browse past news stories and their summaries.
Authors: Narasimhula L V Venugopal, K Visala, Sammingi Nirmala, Ch Vinod Varma, Adibabu Triparagiri, Athmakuri Satish Kumar, Ch Sekhar
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4405920430