Implementation of Support Vector Machine (SVM) Algorithm in Sentiment Analysis of the Mobile JKN Application Using the CRISP-DM Approach

Nurdiyanto Yusuf; Ilham Bintang

doi:10.63956/jitar.v1i2.32

Authors

Nurdiyanto Yusuf Gunadarma University
Ilham Bintang Gunadarma University

DOI:

https://doi.org/10.63956/jitar.v1i2.32

Keywords:

Application, Mobile JKN, Sentiment analysis, Support Vector Machine, Crisp-DM, Google Play Store.

Abstract

Technological advances have facilitated various sectors, including healthcare in Indonesia. One innovation is the Mobile JKN application developed by BPJS Kesehatan to ease access to National Health Insurance (JKN) services. This study aims to analyze user sentiment toward the Mobile JKN application using 1,000 reviews from the Google Play Store. The analysis was carried out using the Support Vector Machine (SVM) method with the Cross-Industry Standard Process for Data Mining (CRISP-DM) approach. The data was collected through scraping and processed through pre-processing stages, TF-IDF, data visualization, and splitting into training and testing sets to classify sentiment into positive and negative categories. The deployment stage was carried out by creating a visualization website using Streamlit, which includes dashboards, word clouds, various chart types, text tables, and a manual sentiment prediction feature based on user input. The website is available at: https://sentimen-mobilejkn-ebd8w9s59ueehw6wjqhoct.streamlit.app/. After all calculations were performed, the results showed that the implementation of the Support Vector Machine (SVM) algorithm to analyze user sentiment of the Mobile JKN application based on reviews from the Google Play Store using a 70% training and 30% testing data split achieved an Accuracy of 86%, Precision of 86%, Recall of 86%, and F1-Score of 87%. These findings may support future development of the Mobile JKN service.

References

Iskandar, J.W. and Natalia, Y., Comparison of Naïve Bayes, SVM, and k-NN for Aspect-Based Sentiment Analysis of Gadgets, JURNAL RESTI (Journal of Systems Engineering and Information Technology), 5(2), pp.1120–1126. DOI: https://doi.org/10.29207/resti.v5i6.3588, 2021.

Singgalen, Y.A., Performance Analysis of NBC, DT, and SVM Algorithms in Classifying Visitor Review Data of Borobudur Temple Based on CRISP-DM, Building of Informatics, Technology and Science (BITS), 4(3), pp.1634–1646. DOI: https://doi.org/10.47065/bits.v4i3.2766, 2022.

Purwanti, Z. and Sugiyono, Text Mining Modeling for Sentiment Analysis of the Free Lunch Program on X Social Media Using the Support Vector Machine (SVM) Algorithm, Jurnal Indonesia: Management Informatics and Communication, 5(3), pp.3065–3077. DOI: https://journal.stmiki.ac.id/index.php/jimik/, 2024.

Munawaroh, A., Ridhoi, R. and Rudiman, Sentiment Analysis Using Naïve Bayes Based on Orange for IKN Development Risk, Jurnal Mahasiswa Teknik Informatika, 8(1), pp.587–592. DOI: https://doi.org/10.36040/jati.v8i1.8454, 2024.

Hastono, W., Aini, N., Karno, A.S.B. and Rere, L.M.R., Machine Learning Methods for Predicting Manure Management Emissions, Jurnal Nasional Teknik Elektro dan Teknologi Informasi, 11(22), pp.2460–5719. DOI: http://dx.doi.org/10.22146/jnteti.v11i2.2586, 2022.