Publication

Abstract : Automatic summarization of news articles plays a crucial role in information retrieval and consumption. This research paper presents an approach for summarizing Google News articles across four distinct classes: business, sports, entertainment, and technology. We collect a set of news items from these categories by utilizing the T5 Transformer model. The acquired articles are preprocessed and turned into a text-to-text format, where the complete article text is the input and a succinct summary is the desired output. The T5 model may be improved on this dataset to produce precise and useful summaries for each news category. Moreover, the generated summaries are accompanied by links that direct users to the original articles, providing an option for accessing the complete content. Experimental evaluation using established metrics demonstrates the effectiveness of our approach in generating high-quality summaries across different news domains. This research contributes to improving information retrieval and consumption by providing users with succinct summaries while retaining the option to explore the full article for more in-depth information.