Automatic Generation of Class Models from Korean Software Requirements 


Vol. 15,  No. 3, pp. 210-218, Mar.  2026
https://doi.org/10.3745/TKIPS.2026.15.3.210


PDF
  Abstract

Software requirements are specifications that define the functionalities and constraints a system must possess, written in natural language, and design work is carried out using standardized UML class diagrams. Although natural language requirements offer flexibility, they are difficult to validate accurately and can be subject to subjective interpretations, leading to consistency issues between requirements and class diagrams. To address this, research on the automatic conversion of natural language requirements into class diagrams has been actively pursued. However, most of this research focuses on requirements written in English, and visualization has not been performed using CASE tools commonly used in the industry. This paper proposes a method to automatically generate class models from Korean software requirements. It takes software requirements written in Korean as input and applies natural language processing techniques and 5W1H-based heuristic rules to extract components of class diagrams. Based on the extracted information, the class diagram is visualized using the model interface language of CASE tools. Additionally, precision, recall, and overgeneration of the developed generator are measured to demonstrate its performance. The results show high performance with a precision of 89%, recall of 80%, and overgeneration of 10%.

  Statistics
  Cite this article

[IEEE Style]

S. Jung and W. Lee, "Automatic Generation of Class Models from Korean Software Requirements," The Transactions of the Korea Information Processing Society, vol. 15, no. 3, pp. 210-218, 2026. DOI: https://doi.org/10.3745/TKIPS.2026.15.3.210.

[ACM Style]

Seungmo Jung and Woojin Lee. 2026. Automatic Generation of Class Models from Korean Software Requirements. The Transactions of the Korea Information Processing Society, 15, 3, (2026), 210-218. DOI: https://doi.org/10.3745/TKIPS.2026.15.3.210.

Forms

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

Old Journals

Indexing

All publications of TKIPS is indexed in DOI, EBSCO, Google Scholar, Crossref, and CrossCheck.

TKIPS is also selected as the Journal for Accreditation by NRF (National Research Foundation of Korea).

Related Journals