Analisis Klasifikasi Sentimen Review Mobile Jkn Menggunakan Naïve Bayes Classifier Dan Support Vector Machine

Romadhona, Melita (2021) Analisis Klasifikasi Sentimen Review Mobile Jkn Menggunakan Naïve Bayes Classifier Dan Support Vector Machine. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text PA_Melita Romadhona_ASLI.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (1MB) | Request a copy

Abstract

Salah satu penangganan pandemi COVID-19 oleh pemerintah adalah mengeluarkan kebijakan pemberlakuan peraturan physical distancing. Dengan adanya kebijakan tersebut mengakibatkan banyak kegiatan layanan yang dilakukan lewat online. Salah satu instansi yang melakukan layanan online adalah BPJS Kesehatan. Bentuk pelayanan online BPJS Kesehatan dapat diakses melalui aplikasi mobile JKN. Layanan aplikasi BPJS yang berbasis online ini menawarkan banyak kemudahan akan tetapi, adanya kemudahan tersebut pasti juga banyak pengguna yang merasa tidak puas akan aplikasi mobile JKN. Oleh karena itu, perlu dilakukan kajian pada aplikasi ini salah satunya yaitu dengan cara menganalisis komentar pada aplikasi mobile JKN. Algoritma yang digunakan pada kajian ini yaitu Naïve Bayes Classifier dan Support Vector Machine. Hasil analisis sentimen menunjukkan bahwa sebanyak 4001 review bersentimen negatif sedangkan sisanya 335 review bersentimen positif. Hasil analisis klasifikasi algoritma Support Vector Machine dengan Kernel RBF kombinasi parameter C dan Gamma sebesar 100 dan 0,01 memiliki nilai AUC tertinggi sebesar 60,81%. Hasil perbandingkan ketepatan klasifikasi algoritma Naïve Bayes Classifier dan Support Vector Machine memiliki nilai AUC berturut-turut 63,91% dan 60,81 % dari nilai AUC tersebut maka algoritma terbaik pada hasil analisis sentimen review mobile JKN menggunakan algoritma Naïve Bayes Classifier.
==================================================================================================================
One of the government's handling of the COVID-19 pandemic is issuing a policy to enforce physical distancing rules. With this policy, many service activities are carried out online. One of the agencies that provide online services is BPJS Health. BPJS Health's online is form of service can be accessed through the JKN mobile application. This service offers many conveniences, but there are many users are not satisfied with the JKN mobile application. Therefore, it is necessary to conduct a study on this application, one of which is analyzing the comment on the JKN mobile application. The algorithms used in this study are Nave Bayes Classifier and Support Vector Machine. The results of analysis show that as many as 4001 comment reviews have a negative sentiment while the remaining 335 comment reviews have a positive sentiment. The results of the classification analysis using Support Vector Machine algorithm with Kernel RBF with a combination of parameters C and Gamma of 100 and 0.01 have the highest AUC value of 60.81%. The results per comparison of the accuracy of the classification of the Naïve Bayes Classifier and Support Vector Machine algorithms have an AUC value of 63.91% and 60.81% of the AUC value, respectively, the best algorithm on the results of JKN mobile sentiment analysis uses the Naïve Bayes Classifier algorithm.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	COVID-19, Naïve Bayes Classifier, Mobile JKN, Support Vector Machine COVID-19, Naïve Bayes Classifier, Mobile JKN, Support Vector Machine
Subjects:	H Social Sciences > HA Statistics
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics
Depositing User:	Melita Romadhona
Date Deposited:	12 Aug 2021 13:05
Last Modified:	12 Aug 2021 13:05
URI:	http://repository.its.ac.id/id/eprint/85874

Actions (login required)

View Item