REVIEW APLIKASI KREDIVO MENGGUNAKAN ANALISIS SENTIMEN DENGAN ALGORITMA SUPPORT VECTOR MACHINE

Abstrak

Kredivo is an instant credit application that makes it easy to buy now and pay later. This application is also known as a digital credit card where borrowers can use it for online instalment shopping. Apart from being used for shopping purposes, borrowers can also cash out in the form of cash. However, not all users are satisfied with this service, there are many comments, one of which is conveyed through the Credivo application review feature on the Google Play Store. Therefore, in this study, researchers tried to do sentiment analysis of the credivo application review on the google play store using the support vector machine algorithm. This study aims to determine the results of sentiment analysis by classifying reviews to be positive or negative and seeing how high the accuracy value is generated by the support vector machine algorithm. From tests conducted on 10000 user review data for the Credivo application on the Google Play store with a division of training and testing data of 80:20, it was found that reviews containing positive sentiment were 59% and containing negative sentiment were 41%. Evaluation of classification results with confusion matrix testing obtained an accuracy value of 87%, a precision value of 84%, a recall value of 85%, and an f1 score value of 85%.