RANCANG BANGUN VS CODE EXTENSION UNTUK IDENTIFIKASI KODE YANG BERISIKO MENGHAMBAT MAINTAINABILITY MENGGUNAKAN SOFTWARE METRICS DAN LARGE LANGUAGE MODEL

Ekasari Amalia, NIM.: 21106050023 (2026) RANCANG BANGUN VS CODE EXTENSION UNTUK IDENTIFIKASI KODE YANG BERISIKO MENGHAMBAT MAINTAINABILITY MENGGUNAKAN SOFTWARE METRICS DAN LARGE LANGUAGE MODEL. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

	Text (RANCANG BANGUN VS CODE EXTENSION UNTUK IDENTIFIKASI KODE YANG BERISIKO MENGHAMBAT MAINTAINABILITY MENGGUNAKAN SOFTWARE METRICS DAN LARGE LANGUAGE MODEL) 21106050023_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (5MB)
	Text (RANCANG BANGUN VS CODE EXTENSION UNTUK IDENTIFIKASI KODE YANG BERISIKO MENGHAMBAT MAINTAINABILITY MENGGUNAKAN SOFTWARE METRICS DAN LARGE LANGUAGE MODEL) 21106050023_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (15MB) | Request a copy

Abstract

Maintainability is a software quality attribute related to the ease of understanding, repairing, modifying, and developing code. This study aims to develop the Maintainability Risk Analyzer as a VS Code Extension to identify maintainability risks in JavaScript code using software metrics and provide contextual explanations through a Large Language Model (LLM). The system was developed using Extreme Programming (XP), consisting of planning, design, coding, and testing. Static analysis utilizes the Abstract Syntax Tree (AST) to obtain Halstead Volume, Cyclomatic Complexity, and Lines of Code (LOC), which are used to calculate the Visual Studio Maintainability Index (MIVS) and determine the risk level. The analysis results are displayed directly within the development environment through Diagnostic, Highlight, Hover, and Show Summary. The Explain Maintainability Risk feature provides contextual explanations of risk factors and LLM-based refactoring suggestions. The implementation and testing results show that the extension can perform maintainability analysis on JavaScript code, display risk levels and the corresponding metrics, and provide explanations and improvement suggestions through the LLM-based feature. User evaluation of the LLM-based feature using the Extended TAM among 101 valid respondents resulted in an overall mean score of 3.49, which falls into the very high category.

Share this knowledge with your friends :

Actions (login required)

View Item