The ability of computers to read lengthy materials and produce concise summaries that highlight the key ideas is examined in this research. We describe two primary strategies: extractive methods, in which computers identify significant sentences, and abstractive methods, in which computers generate new sentences that encompass crucial concepts. The summaries produced by recent AI models such as BART, T5, and PEGASUS sound far more human. We experimented with these various methods and discovered that although more recent models produce summaries that sound more realistic, they occasionally contain factual errors. We also talk about the problems that still need to be resolved and how these tools benefit people in a variety of fields, including study, media, and law.
Volume 17 | Issue 1
Pages: 93-95