Analisis Sentimen dan Monitoring Kualitas Layanan Mobile JKN Menggunakan Metode Support Vector Machine dan Peta Kendali P

Rafika, Desy Nazalia (2024) Analisis Sentimen dan Monitoring Kualitas Layanan Mobile JKN Menggunakan Metode Support Vector Machine dan Peta Kendali P. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2043201015-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2026. Download (2MB) | Request a copy

Abstract

Pada saat ini berbagai aspek kehidupan manusia tidak dapat dipisahkan dengan teknologi berbasis digital termasuk aspek pelayanan publik salah satunya yaitu Badan Penyelenggara Jaminan Sosial (BPJS) Kesehatan. BPJS mulai memperkenalkan aplikasi layanan berbasis online yang biasa disebut dengan Mobile JKN (Jaminan Kesehatan Nasional). Mobile JKN menangani administrasi hingga layanan kesehatan yang bisa diakses langsung melalui smartphone tanpa perlu repot datang dan antri di kantor BPJS. Tetapi pada Mei 2021 aplikasi ini mengalami peretasan dan kebocoran data. Hal tersebut banyak menimbulkan opini negatif dari masyarakat terhadap aplikasi Mobile JKN. Sehingga akan dilakukan analisis sentimen dan monitoring kualitas layanan dengan metode Support Vector Machine (SVM) dan peta kendali p. Berdasarkan hasil analisis diperoleh proporsi sentimen negatif cenderung lebih besar daripada sentimen positif. Hasil klasifikasi menggunakan SVM diperoleh nilai cost dan gamma terbaik yaitu 1 dan 0,014 dengan hasil akurasi data training dan testing masing-masing sebesar 95% dan 96% masuk dalam kategori klasifikasi sangat baik. Hasil monitoring control chart dengan peta kendali p pada data testing aktual dan prediksi didapatkan sentimen negatif ulasan pengguna aplikasi Mobile JKN pada Google Play Store terkendali secara statistik. Situasi terkendali menandakan bahwa pengelolaan sentimen negatif pada ulasan pengguna aplikasi Mobile JKN dapat dianggap sebagai sesuatu yang teratur dan dapat diprediksi, bukan disebabkan oleh faktor-faktor yang tidak terkendali. Sehingga dapat disimpulkan bahwa kualitas layanan aplikasi Mobile JKN pada Google Play Store cukup baik dan membuktikan bahwa hasil klasifikasi menggunakan metode SVM dengan kernel RBF yaitu akurat atau sangat baik.
=================================================================================================================================
At this time, various aspects of human life cannot be separated from digital-based technology, including aspects of public services, one of which is the Social Security Organizing Agency (BPJS) Health. BPJS began to introduce an online-based service application commonly called Mobile JKN (National Health Insurance). Mobile JKN handles administration to health services that can be accessed directly via smartphone without the hassle of coming and queuing at the BPJS office. But in May 2021 this app suffered a hack and data leak. This has caused a lot of negative opinions from the public towards the JKN Mobile application. So that sentiment analysis and service quality monitoring will be carried out using the Support Vector Machine (SVM) method and control map p. Based on the results of the analysis, the proportion of negative sentiment tends to be greater than positive sentiment. The classification results using SVM obtained the best cost and gamma values of 1 and 0.014 with the results of training and testing data accuracy of 95% and 96% respectively included in the very good classification category. The results of monitoring control charts with control p chart on actual testing data and predictions obtained negative sentiment of JKN Mobile application user reviews on the Google Play Store controlled statistically. The controlled situation indicates that the management of negative sentiment in user reviews of the JKN Mobile application can be considered as something orderly and predictable, not caused by uncontrollable factors. So it can be concluded that the quality of JKN Mobile application services on the Google Play Store is quite good and proves that the classification results using the SVM method with the RBF kernel are accurate or very good.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Google Play Store, Mobile JKN, Peta Kendali P, Support Vector Machine, Google Play Store
Subjects:	Q Science Q Science > Q Science (General) Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Desy Nazalia Rafika
Date Deposited:	20 May 2024 04:55
Last Modified:	20 May 2024 04:55
URI:	http://repository.its.ac.id/id/eprint/107986

Actions (login required)

View Item