Putri, Ramadhana Candra Kirana Aditya (2023) Implementasi Particle Swarm Optimization Pada Analisis Sentimen Ulasan Aplikasi Jaminan Kesehatan Nasional (JKN Mobile) Menggunakan Algoritma Support Vector Machine. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 06311940000047-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
06311940000047-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy |
Abstract
Pengguna internet di Indonesia setiap tahun terus meningkat. Menurut data dari Badan Pusat Statistik pada tahun 2021 sebanyak 98,7% masyarakat yang mengakses internet lebih memilih menggunakan telepon seluler daripada perangkat lain. Karena kepopuleran telepon seluler terus meningkat, maka muncul layanan bernama m-health. Layanan m-health merupakan layanan medis dan kesehatan masyarakat yang dapat diakses melalui ponsel. BPJS Kesehatan sebagai penyelenggara Jaminan Kesehatan Nasional yang ditunjuk oleh Pemerintah Indonesia juga berupaya untuk meningkatkan kualitas pelayanan dan kemudahan aksesibilitas kesehatan melalui Layanan m-health. Maka dari itu, BPJS Kesehatan meluncurkan aplikasi Jaminan Kesehatan Nasional (JKN Mobile). Untuk melihat kualitas dan kepuasan pengguna terhadap aplikasi ini dapat menggunakan analisis sentimen melalui ulasan yang telah diberikan. Salah satu algoritma dapat digunakan untuk menganalisis sentimen pengguna adalah dengan menggunakan Support Vector Machine. Namun karena Support Vector Machine mempunyai banyak atribut yang digunakan, diperlukan suatu algoritma lain yang berfungsi sebagai seleksi fitur, maka dari itu dipilihlah seleksi fitur menggunakan Particle Swarm Optimization. Data yang digunakan berupa data ulasan pengguna JKN Mobile di Google Play Store. Dari data tersebut akan dibagi menjadi dua kelas sentimen, yaitu positif dan negatif. Selanjutnya data akan diklasifikasi menggunakan Support Vector Machine dan Support Vector Machine menggunakan Particle Swarm Optimization. Hasil penelitian menunjukkan bahwa dengan adanya seleksi fitur Particle Swarm Optimization, nilai accuracy Support Vector Machine meningkat. Untuk model paling baik adalah SVM Kernel RBF dengan PSO dengan accuracy sebesar 92,39%, precision sebesar 80,83%, recall sebesar 85,09%, F1-Score sebesar 82,91%, dan AUC sebesar 89,75%.
=================================================================================================================================
Internet users in Indonesia continue to increase every year. According to data from Badan Pusat Statistik, in 2021 as many as 98.7% of people who access the internet will prefer to use cell phones over other devices. Because the popularity of cell phones continues to increase, a program called m-health appears. The m-health service is a medical and public health service that can be accessed via cellphone. BPJS Kesehatan as the organizer of the National Health Insurance appointed by the Government of Indonesia is also trying to improve the quality of service and easy accessibility of health through m-health services. Therefore, BPJS Kesehatan launched the National Health Insurance application (JKN Mobile). To see the quality and user satisfaction with this application, sentiment analysis can be used through the reviews that have been given. One of the algorithms that can be used to analyze user sentiment is a Support Vector Machine. However, because the Support Vector Machine has many attributes that are used, another algorithm is needed to function as feature selection, therefore feature selection was chosen using Particle Swarm Optimization. The data used is JKN Mobile user review data on the Google Play Store. The data will be divided into two sentiment classes, namely positive and negative. Furthermore, the data will be classified using Support Vector Machine and Support Vector Machine using Particle Swarm Optimization. The results showed that with the Particle Swarm Optimization feature selection, the accuracy value of the Support Vector Machine increased. The best model is SVM Kernel RBF with PSO with an accuracy of 92.39%, precision of 80.83%, recall of 85.09%, F1-Score of 82.91%, and AUC of 89.75%.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Analisis Sentimen, JKN Mobile, Particle Swarm Optimization, Support Vector Machine, Ulasan Aplikasi, App Review, Sentiment Analysis |
| Subjects: | Q Science > QA Mathematics > QA401 Mathematical models. Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis |
| Depositing User: | Ramadhana Candra Kirana Aditya Putri |
| Date Deposited: | 12 Jul 2023 06:21 |
| Last Modified: | 12 Jul 2023 06:21 |
| URI: | http://repository.its.ac.id/id/eprint/98423 |
Actions (login required)
 |
View Item |