Kurniawan, Hafiz (2024) VISUALISASI KERUNUTAN ANTARA KALIMAT KEBUTUHAN FUNGSIONAL DAN CODING DENGAN MENGGUNAKAN GRAF. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of TA_5025201032.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
TA_5025201032.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (3MB) | Request a copy |
Abstract
Keberhasilan pengembangan perangkat lunak diukur berdasarkan kemampuan aplikasi dalam memenuhi kebutuhan pengguna. Untuk melacak kesesuaian kebutuhan dengan implementasi dapat menggunakan artefak kebutuhan fungsional dan implementasi coding yang dilakukan secara manual. Namun, metode manual memiliki kekurangan dalam konsumsi waktu yang lama, rawan kesalahan, serta membutuhkan upaya yang besar. Untuk menangani kekurangan dalam metode pengecekan secara manual, pada Tugas Akhir ini akan dikembangkan kakas bantu yang dapat membantu mengecek kesesuaian antara kalimat kebutuhan fungsional dan implementasi coding pada perangkat lunak. Artefak kalimat kebutuhan fungsional dan source code akan diproses untuk mendapatkan properti-properti penting yang terkandung di dalamnya. Natural Languange Processing (NLP) digunakan pada kakas bantu untuk mengidentifikasi kerunutan antara kalimat kebutuhan fungsional dan source code yang meliputi tahap preprocessing dan perhitungan kemiripan (similarity). Tahap preprocessing yang dilakukan berupa tokenisasi, lowercasing, dependency parsing, lemmatization, stop words removal, dan normalisasi. Tahap perhitungan kemiripan (similarity) dengan menggunakan kesamaan semantik berbasis Path Similarity. Hasil dari identifikasi kemiripan tertinggi akan menjadi tautan jejak (trace link) dari kalimat kebutuhan fungsional. Kerunutan direpresentasikan dengan graf yang memvisualisasikan kerunutan dari kalimat kebutuhan fungsional ke source code. Kerunutan digambarkan dengan simpul terhubung antara simpul kalimat kebutuhan fungsional ke simpul source code. Kakas bantu diujicobakan pada proyek Monthes, yang dikembangkan dengan arsitektur Model View Controller (MVC). Berdasarkan pengujian yang dilakukan didapatkan nilai accuracy sebesar 0.768, nilai precision sebesar 0.954, nilai recall sebesar 0.797 dan nilai F-measure sebesar 0.868.
========================================================================================================================
The success of software development is measured based on the application's ability to meet user needs. To track the conformity of requirements with implementation, functional requirement artifacts and coding implementation can be done manually. However, the manual method has shortcomings in consuming a long time, prone to errors, and requires a lot of effort. To handle the shortcomings in the manual checking method, this Final Project will develop a tool that can help check the conformity between functional requirement statement and coding implementation in software. Artifacts of functional requirement statements and source code will be processed to obtain important properties contained in them. Natural Language Processing (NLP) is used in the tool to identify the harmony between functional requirement statements and source code, which includes preprocessing and similarity calculation. The preprocessing stage includes tokenization, lowercasing, dependency parsing, lemmatization, stop words removal, and normalization. The similarity calculation stage uses semantic similarity based on Path Similarity. The result of the highest similarity identification will be the trace link of the functional requirement statement. The trace link is represented with a graph model that visualizes the trace link from the functional requirement statement to the source code. The concatenation is depicted with connected nodes between functional requirement statement nodes to source code nodes. The tool tested on the Monthes project, which was developed with Model View Controller (MVC) architecture. Based on the testing, the accuracy value is 0.768, the precision value is 0.954, the recall value is 0.797 and the F-measure value is 0.868.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Graf, Traceability, Visualisasi, Graph, Traceability, Visualization. |
| Subjects: | Q Science > QA Mathematics > QA76.9.I52 Information visualization |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Hafiz Kurniawan |
| Date Deposited: | 08 Aug 2024 03:00 |
| Last Modified: | 08 Aug 2024 03:00 |
| URI: | http://repository.its.ac.id/id/eprint/111529 |
Actions (login required)
 |
View Item |