Scholarlink

This chapter introduces the use of generative large language models (LLMs) in natural language analysis (NLA), focusing on their potential to extract soft data from complex textual information for various analytical purposes. It sets the stage for the book as a practical guide to the application of LLMs in NLA, particularly using open source models on a local system. Given the typical use of LLMs in translation and summarization, it emphasizes the need for a transdisciplinary approach that integrates knowledge from fields such as cognitive semiotics, linguistics, computer science and big data in order to be applied to analytics. Finally, the chapter uses the metaphor of LLMs as “language calculators” to illustrate their function in generating meaningful outputs through semantic embeddings, while also addressing the inherent variability and limitations of these models.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Introduction

  • 章节

  • 发表日期：

    2025年

查看全文
  • Francisco S. Marcondes,
  • Adelino Gala,
  • Renata Magalhães,
  • Fernando Perez de Britto,
  • Dalila Durães,
  • Paulo Novais

摘要

This chapter introduces the use of generative large language models (LLMs) in natural language analysis (NLA), focusing on their potential to extract soft data from complex textual information for various analytical purposes. It sets the stage for the book as a practical guide to the application of LLMs in NLA, particularly using open source models on a local system. Given the typical use of LLMs in translation and summarization, it emphasizes the need for a transdisciplinary approach that integrates knowledge from fields such as cognitive semiotics, linguistics, computer science and big data in order to be applied to analytics. Finally, the chapter uses the metaphor of LLMs as “language calculators” to illustrate their function in generating meaningful outputs through semantic embeddings, while also addressing the inherent variability and limitations of these models.