Extraction of Step Performed in Use Case Description as a Reference for Conformity of Sequence Diagrams Using Text Mining (Case Study: SRS APTU)

Abstract

Extraction is an essential part of processing a document to ensure the success of the text mining process. In this study, the example of the SRS document used is the Integrated Service Application (APTU) KPKNL Bandung, an application to manage the process of submitting service tickets at the State Property and Auction Service Office. There is a difference in interpreting the activities that exist in the Use Case Description artifact with a Sequence Diagram that provides an overview of the functionality of a process to show the involvement of an activity related to the Use Case Description. This study aims to perform step extraction on the Use Case description. The results of this extraction are compared for their suitability with the sequence diagram using the concept of text mining. There are core results from this research activity. First, the highest similarity between documents is in the SP01 and SD01 documents, with the similarity value being 0.69108792. Second, the highest similarity between words is found in words “list” and “menu,” with the similarity value being 0.9412. Third, the Kappa Score from Gwet's AC1 formula using the Python programming language is 0.12362, which means “Slight Agreement,” while the Kappa Score value using a questionnaire filled in by the expert is 0.97464, which means “Almost perfect.