본문 바로가기
  • Home

A Lifecycle-Based Integrated Modeling Framework for Generative AI Services

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2026, 31(9), pp.125~133
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : June 4, 2026
  • Accepted : September 5, 2026
  • Published : September 30, 2026

Eun-Sook Cho 1

1서일대학교

Accredited

ABSTRACT

Generative AI services require lifecycle support for Prompt Engineering, Retrieval-Augmented Generation (RAG), user experience (UX), and continuous improvement, which existing frameworks such as CRISP-DM, OSEMN, MLOps, and CRISP-ML(Q) do not fully address. This paper compares these frameworks using lifecycle-based criteria and proposes an integrated framework that incorporates business understanding, UX, data and RAG design, Prompt Engineering, generative AI modeling, service design, LLMOps-based operation, monitoring and feedback, and continuous improvement. Comparative analysis and exploratory expert evaluation indicate the conceptual applicability and structural adequacy of the proposed framework for supporting the Generative AI service lifecycle. The results are exploratory and do not constitute empirical evidence of performance improvement.

Journal Copyright Policy

No CCL information provided

Citation status

* References for papers published after 2025 are currently being built.