Sentimen Analisis Penggunaan Aplikasi Canva Menggunakan Support Vector Classification

Abstract

This Canva graphic design application is very popular among many people including students, where this application is very helpful in developing graphic ideas with features that are easy to use by users to create an attractive design and make time efficient. This study discusses the classification of sentiment towards Canva applications using the Naïve Bayes and Support Vector Machine methods. In this study, the data used are user reviews taken from the Google Play Store application, then the reviews are analyzed and classified into three categories, namely, positive, negative, neutral and after that the data will be processed in several stages; data collection to research results. And the result of this study is that the Support Vector Machine (SVM) method which has the best performance with SVC parameters gets an accuracy of 77.48%, followed by Support Vector Machines (SVM) with LinearSVC parameters with an average accuracy value of 71.80%.