Sistem Pakar Diagnosis Gangguan Tidur Pada Mahasiswa Akibat Penggunaan Gadget Berbasis Forward Chaining Dan Certainty Factor

Aminah, Nur (2026) Sistem Pakar Diagnosis Gangguan Tidur Pada Mahasiswa Akibat Penggunaan Gadget Berbasis Forward Chaining Dan Certainty Factor. Skripsi thesis, Universitas Islam Negeri Imam Bonjol Padang.

	Text (Cover - Daftar Isi) FILE Cover - Daftar Isi.pdf - Published Version Download (716kB)
	Text (BAB I) FILE BAB I.pdf - Published Version Download (383kB)
	Text (BAB III) FILE BAB III.pdf - Published Version Download (584kB)
	Text (BAB V & Daftar Pustaka) FILE BAB V & Daftar Pustaka.pdf - Published Version Download (397kB)
	Text (Skripsi FULL TEXT) Skripsi FULL TEXT.pdf - Published Version Restricted to Repository staff only Download (3MB)

Abstract

Excessive gadget use, particularly before bedtime, has been shown to negatively affect students' sleep quality due to blue light exposure that suppresses melatonin production. The identification of sleep disorders among students at Universitas Islam Negeri Imam Bonjol Padang is currently constrained by limited access to professional diagnostic services, while existing preliminary screening methods have not integrated sleep disorder symptoms with gadget-use behavior patterns. This study aims to design and build an expert system for diagnosing sleep disorders in students caused by gadget use, using the Forward Chaining and Certainty Factor (CF) methods, and to test the system's accuracy using a Confusion Matrix. The study used the Research and Development (R&D) method with a Waterfall development model limited to the construction stage, covering communication, planning, modeling, coding, and system testing. The system's knowledge base covers 10 types of sleep disorders along with 57 symptom codes (G01-G57), including four gadget-use behavior indicators (G54-G57), compiled through a literature study and validated by a clinical psychology expert. The system was developed as a web application using PHP and MySQL, in which the Forward Chaining method serves as the inference engine that traces the symptoms and gadget-use behaviors selected by the user toward a diagnostic conclusion, while Certainty Factor calculates the diagnosis confidence level as a percentage. System testing was carried out in two stages, namely Black Box Testing to ensure that all system functions operate as designed, and accuracy testing by comparing the system's screening results with 15 test cases directly validated by the psychology expert using a one-vs-rest Confusion Matrix approach. The test results showed that all system diagnoses, covering six types of sleep disorders identified in the test data, matched the expert's assessment with no False Positive or False Negative cases found, resulting in a system accuracy of 100%. However, these results cannot yet be generalized broadly given the limited size and variation of the test data and validation involving only a single expert. These findings indicate that the developed expert system is a feasible tool for the preliminary screening of gadget-related sleep disorders among students, though it is not intended to replace clinical diagnosis by professional practitioners.

Actions (login required)

View Item