Traditionally, early stages of the Software Development Life Cycle (SDLC), such as requirement elicitation and analysis are manual, require significant expertise. Creation of purposive machine-processable models is an intellect-intensive task requiring business domain as well as modeling expertise, which is in short supply. To address this, we propose the LLM4Model tool, a configurable, extensible, domain-agnostic solution for automating model authoring with Large Language Models (LLMs) to reduce cognitive load. We demonstrate its effectiveness with two real-world case studies, showing automated authoring of requirements specification model and sharing the results.

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

LLM4Model: Automated Requirements Specification Model Authoring

  • Asha Rajbhoj,
  • Akanksha Somase,
  • Tanay Sant,
  • Sushant Vale,
  • Vinay Kulkarni

摘要

Traditionally, early stages of the Software Development Life Cycle (SDLC), such as requirement elicitation and analysis are manual, require significant expertise. Creation of purposive machine-processable models is an intellect-intensive task requiring business domain as well as modeling expertise, which is in short supply. To address this, we propose the LLM4Model tool, a configurable, extensible, domain-agnostic solution for automating model authoring with Large Language Models (LLMs) to reduce cognitive load. We demonstrate its effectiveness with two real-world case studies, showing automated authoring of requirements specification model and sharing the results.