From Extractive to Generative: An Analysis of Automatic Text Summarization Techniques

School of Artificial Intelligence, Tiangong University, 300000, Tianjin, China

^* Corresponding author: 2013810112@tiangong.edu.cn

Abstract

With the explosive growth of digital content, the demand for effective information retrieval and summarization has become increasingly important. This paper provides a comprehensive review of automated text summarization techniques, focusing on the challenge of condensing large volumes of text into concise summaries. The article explores the evolution of automatic summarization methods, from early extractive techniques to modern generative approaches based on deep learning. The review highlights significant milestones in the development of summarization algorithms, including the emergence of Transformer-based models like Bidirectional Encoder Representations from Transformers (BERT) and Generative Pre-trained Transformer (GPT), which have significantly improved the quality and coherence of generated summaries. Additionally, the paper examines the diverse applications of summarization technologies across fields such as healthcare, discussing the challenges and solutions presented in the literature. By shedding light on the current advancements and ongoing challenges, this review underscores the crucial role of automated text summarization in enhancing information accessibility, with promising implications for future research and practical applications.