Identifikasi Kesesuaian Ubiquitous Language dalam Domain-Driven Design Menggunakan Traceability pada Studi Kasus myITS Thesis

Giovanni, Decya (2026) Identifikasi Kesesuaian Ubiquitous Language dalam Domain-Driven Design Menggunakan Traceability pada Studi Kasus myITS Thesis. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025221027-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Ubiquitous language merupakan konsep kunci dalam Domain-Driven Design (DDD) untuk menjaga keselarasan pemahaman antara kebutuhan bisnis dan implementasi sistem, namun dalam praktiknya sering terjadi ketidaksesuaian istilah antara user story dan kode sumber. Tugas Akhir ini mengembangkan metode software traceability otomatis untuk mengidentifikasi dan mengukur kesesuaian ubiquitous language pada sistem berarsitektur Clean Architecture yang menerapkan paradigma DDD. Metode yang diusulkan memanfaatkan Property Listing Task (PLT) untuk mengekstraksi fitur artefak dan dikombinasikan dengan teknik Natural Language Processing (NLP) dan Information Retrieval (IR). Proses ini meliputi preprocessing teks, representasi vektor menggunakan TF-IDF dan FastText, serta pengukuran similaritas melalui pendekatan yang menggabungkan cosine similarity, exact word match, dan property-based similarity. Analisis traceability dilakukan dengan mempertimbangkan bounded context agar kesesuaian istilah domain dievaluasi pada ruang lingkup modul yang tepat. Evaluasi dilakukan pada proyek myITS Thesis dengan membandingkan kondisi kode sumber As-Is dan Modified. Hasil pengujian menunjukkan bahwa konfigurasi P3 Modified memperoleh performa terbaik dengan precision 0,5242, recall 0,6771, F1-score 0,5909, dan accuracy 0,9261. Dibandingkan P3 As-Is, F1-score meningkat dari 0,5047 menjadi 0,5909, sedangkan kategori ubiquitous language conformant meningkat dari 153 menjadi 315 konsep. Hasil tersebut menunjukkan bahwa pendekatan traceability dapat digunakan untuk mengidentifikasi kesesuaian ubiquitous language antara user story, acceptance criteria, dan kode sumber pada sistem berbasis DDD.
=====================================================================================================================================
Ubiquitous language is a key concept in Domain-Driven Design (DDD) that aims to maintain alignment between business requirements and system implementation. However, in practice, inconsistencies in terminology often occur between user stories and source code. This thesis proposes an automated software traceability method to identify and measure the alignment of ubiquitous language in systems adopting Clean Architecture and the DDD paradigm. The proposed method utilizes the Property Listing Task (PLT) to extract artifact features and integrates Natural Language Processing (NLP) and Information Retrieval (IR) techniques. The process includes text preprocessing, vector representation using TF-IDF and FastText, and similarity measurement through a hybrid approach combining cosine similarity, exact word matching, and property-based similarity. Traceability analysis incorporates bounded context to ensure that domain terminology consistency is evaluated within the appropriate module scope. The evaluation was conducted on the myITS Thesis project by comparing the As-Is and Modified source code. The results show that the P3 Modified configuration achieved the best performance, with a precision of 0.5242, recall of 0.6771, F1-score of 0.5909, and accuracy of 0.9261. Compared with P3 As-Is, the F1-score increased from 0.5047 to 0.5909, while ubiquitous language conformant concepts increased from 153 to 315. These findings indicate that traceability can be used to identify ubiquitous language alignment between user stories, acceptance criteria, and source code in a DDD-based system.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	DDD, Ubiquitous Language, Software Traceability, PLT, NLP, IR, DDD, Ubiquitous Language, Software Traceability, PLT, NLP, IR
Subjects:	Q Science > QA Mathematics > QA76.758 Software engineering
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Giovanni Decya
Date Deposited:	24 Jul 2026 07:20
Last Modified:	24 Jul 2026 07:20
URI:	http://repository.its.ac.id/id/eprint/137201

Actions (login required)

View Item