Hisyam, Khariza Azmi Alfajira (2024) Kerunutan Perangkat Lunak Mundur Antara Kalimat Kebutuhan Dengan Kode Sumber Pada Aplikasi Teman Tani. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
5025201044-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (6MB) | Request a copy |
Abstract
Penelitian kerunutan perangkat lunak ini bertujuan untuk mengembangkan metode kerunutan antara kalimat kebutuhan (user story) dengan kode sumber (source code) pada aplikasi TemanTani. Penelitian ini dapat membantu kegiatan software quality assurance dan memaksimalkan metode pencarian kerunutan. Selain itu juga dapat memudahkan pengembang dalam melacak kebutuhan agar mempermudah pemeliharaan perangkat lunak di masa mendatang. Metode yang digunakan pada penelitian ini adalah pencarian kerunutan berbasis semantik yang mengkombinasikan word embedding dan rule dan mengambil studi kasus aplikasi TemanTani yang berarsitektur clean dan microservices serta menggunakan paradigma Domain Driven Design (DDD). Aplikasi TemanTani yang dianalisis terdiri dari empat servis yaitu servis land, user, project, dan investment. Beberapa metode yang diujikan pada aplikasi TemanTani selain word embedding dan rule adalah word embedding tanpa pendekatan aturan, BERT, dan LSTM/GRU. Hasil penelitian ini menunjukkan bahwa metode pencarian kerunutan menggunakan metode word embedding dan rule paling baik digunakan pada servis user. Sedangkan untuk metode lain seperti word embedding saja, BERT, LSTM/GRU memiliki performa terbaik pada servis investment. Metode LSTM/GRU dapat mencapai nilai recall yang tinggi pada semua servis, namun nilai akurasi dan presisi yang lebih tinggi ada pada metode BERT, kecuali pada servis user yang lebih unggul dengan metode word embedding dan rule.
====================================================================================================================================
This software continuity research aims to develop a method of continuity between user stories and source code in the TemanTani application. This research can help software quality assurance activities and maximize traceability search methods. Apart from that, it can also make it easier for developers to track requirements to make software maintenance easier in the future. The method used in this research is a semantic-based sequence search that combines word embedding and rules and takes a case study of the TemanTani application which has a clean architecture and microservices and uses the Domain Driven Design (DDD) paradigm. The TemanTani application analyzed consists of four services, namely land, user, project and investment services. Several methods tested on the TemanTani application apart from word embedding and rules are word embedding without rules approach, BERT, and LSTM/GRU. The results of this research indicate that the sequence search method using word embedding and rule methods is best used for user services. Meanwhile, other methods such as word embedding only, BERT, LSTM/GRU have the best performance in investment services. The LSTM/GRU method can achieve high recall values for all services, but the accuracy and precision values are higher for the BERT method, except for user services which are superior to the word embedding and rule methods.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Domain Driven Design (DDD), Kerunutan, Microservices, TemanTani, Word Embedding; Domain Driven Design (DDD), Microservices, TemanTani, Traceability, Word Embedding |
Subjects: | Q Science > QA Mathematics > QA76.758 Software engineering Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
Depositing User: | Khariza Azmi Alfajira Hisyam |
Date Deposited: | 29 Jul 2024 03:51 |
Last Modified: | 29 Jul 2024 03:51 |
URI: | http://repository.its.ac.id/id/eprint/109227 |
Actions (login required)
View Item |