Sistem Penilaian Jawaban Esai Otomatis Menggunakan Laravel dan API OpenAI

Purnama, I Made Raditya (2025) Sistem Penilaian Jawaban Esai Otomatis Menggunakan Laravel dan API OpenAI. Undergraduate thesis, Politeknik Neger Bali.

	Text (Full Text) RAMA_58302_2115354072_full.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy
	Text (Cover, Bab 1, Bab 5, dan Referensi) RAMA_58302_2115354072_0805098101_0625048702_part.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial Share Alike. Download (549kB)
	Text (iThenticate) RAMA_58302_iThenticate.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Manual essay assessment in educational settings often requires significant time and effort and is susceptible to evaluator subjectivity. This research aims to develop an automatic essay scoring system based on a web application using the Laravel framework integrated with the OpenAI API. The system is designed to assist lecturers or teachers in evaluating essay answers quickly, consistently, and objectively. The system development follows the Waterfall methodology, consisting of requirement analysis, design, implementation, and testing phases. Essay answers submitted by students are automatically assessed by the GPT-3.5 model from OpenAI, which compares the student's response with the ideal answer provided by the instructor. The system also generates a score and constructive feedback for each submission. The testing results show that all system features function as expected, including multi- role login (admin, lecturer, student), essay question management, student answer submission, and automated feedback and scoring storage. Reliability testing using the Mean Absolute Deviation (MAD) method produced a value of 0.6647, indicating a high level of scoring consistency. In conclusion, the system successfully automates essay assessments with accurate and consistent results. It enhances the efficiency and transparency of the evaluation process and provides students with a more interactive and informative learning experience.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Automated Scoring, Essay, Laravel, OpenAI, GPT, Education, Artificial Intelligence.
Subjects:	Ilmu Teknik > Teknik Elektro Dan Informatika > Teknologi Informasi
Divisions:	Jurusan Teknologi Informasi > Prodi D4 Teknologi Rekayasa Perangkat Lunak > Skripsi
Depositing User:	I Made Raditya Purnama
Date Deposited:	15 Sep 2025 05:52
Last Modified:	15 Sep 2025 05:52
URI:	https://repository.pnb.ac.id/id/eprint/20855

Actions (login required)

View Item