WILDANI PUTRI, -
(2024)
ANALISIS SENTIMEN APLIKASI IBADAH ISLAMI PADA GOOGLE PLAY STORE MENGGUNAKAN NAI¨VE BAYERS CLASSIFIER DAN SUPPORT VECTOR MACHINE.
International Symposium On Information Technology And Digital Innovations (ISITDI).
Abstract
This research addresses the rapid increase in adherents of Islam which is directly proportional to the rapid development of technology so that of course Muslims need an application that can help with daily worship activities, one of which is the Muslim Pro application. Muslim Pro has to take user feedback into consideration in order to enhance the application's functionality. It is important to note that the majority of negative review requests will occur in 2023 and 2024. Some of the customer issues that can be found in the reviews section include a lot of annoying ads, frequent error- prone prayer alerts, accounts that are automatically logged out, upgrades that are increasingly problematic, and so forth. Sentiment analysis of user evaluations of the Muslim Pro app on the Google Play Store is the primary goal of this study. In order to categorize user opinions into three groups: negative, neutral, and positive. this study applies the Naïve Bayers Classifier (NBC) and Support Vector Machine (SVM) algorithms on a dataset of 995 reviews. The SMOTE technique is also used for data sampling. When K-Fold Cross Validation with k–10 is applied to enhance the validation findings, the SVM algorithm outperforms the NBC method, achieving an accuracy of 78% compared to the NBC algorithm's 69%. In addition to offering insights on how users see the Muslim Pro application, this study assesses the performance of the NBC and SVM sentiment analysis algorithms.
Actions (login required)
 |
View Item |
