Text Mining and Information Retrieval Optimization of Large Language Models in Digital Libraries

Abstract

Digital libraries are key knowledge management and dissemination platforms in the information age, providing a large amount of literature resources that are crucial for academic research and daily information acquisition. But with the increase of data volume, effectively mining information and improving retrieval efficiency have become challenges. Although traditional text mining and information retrieval techniques have some effectiveness, they still have shortcomings in semantic understanding and complex query processing. In recent years, the development of natural language processing technology, especially large language models (LLMs), has significantly improved text processing capabilities. LLM performs excellently in semantic understanding, text generation, and knowledge extraction through deep learning and large-scale pre training, with strong generalization ability. This article explores how to apply big language models to improve text mining and information retrieval techniques in digital libraries, in order to enhance user retrieval experience and achieve intelligent and personalized information services.