Skripsi

ANALISIS KEPUASAN PENGGUNA PADA PLATFORM E-HEALTH MENGGUNAKAN ENSEMBLE LEARNING DAN ANALISIS TEMATIK

---

The SATUSEHAT application, developed by Indonesia’s Ministry of Health, is a vital digital public service that combines health and technology to promote national development. To assess its impact, this study evaluates public satisfaction with SATUSEHAT by analyzing over 30,000 user feedback from the Google Play Store. Specifically, the research seeks to compare the effectiveness of single and ensemble machine learning models in identifying user sentiment and determining the key variables influencing user satisfaction. Results demonstrate that the Stacking Ensemble Learning Model achieved the highest performance, with 85.9% accuracy, an 85.9% F1-Score, and an AUC-ROC of 0.934. Furthermore, sentiment analysis using the best- performing model found that 75.9% of user reviews expressed negative sentiment. Through thematic analysis using N-gram methods, the study identified the primary themes that influence user satisfaction. Negative themes were then mapped to ISO/IEC 25010 product quality aspects. Notably, the majority of concerns (80.9%) were related to functional suitability and reliability. Other issues included performance efficiency, interaction capability, compatibility, and maintainability. Based on these findings, recommendations are provided for the Ministry of Health and developers to address negative reviews and improve user satisfaction with the application.

---

Availability

Detail Information

Series Title

-

Call Number

T1893722025

Publisher

Indralaya : Prodi Sistem Informasi, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xvii, 85 hlm.; ilus.; tab, 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

005.307

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Informasi
Analisis Tematik--Esemble Learning

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment