-, SERSUSINGSUPPORTVECTORMACHINEANDNAIVE BAYES CLASSIFIES ALGORITH (2024) SENTIMENT ANALYSIS E-WALLET AND DIGITAL BANKING USERS USING SUPPORT VECTOR MACHINE AND NAIVE BAYES CLASSIFIES ALGORITHM. Skripsi thesis, Universitas Islam Negeri Sultan Syarif Kasim Riau.
|
Text
Daffa Takratama Savra - Skripsi (1).pdf Download (8MB) | Preview |
|
![]() |
Text (BAB IV)
bab 4.pdf - Published Version Restricted to Repository staff only Download (1MB) |
Abstract
Digital development in Indonesia is driving the adoption of cashless transactions through e-Wallets and Digital Banking, such as Ovo, Dana, SeaBank, Jenius, and Jago. These payment methods offer advantages like reducing the risk of money loss and theft. To choose the best application, people can look at ratings and reviews on the Google Play Store. However, with the abundance of reviews, sentiment analysis is necessary. This research uses a lexicon based method and evaluation with Support Vector Machine and Naive Bayes Classifier algorithms. The analysis results show that the accuracy of training data and Support Vector Machine for Ovo achieved 95% and 88%, Dana achieved 96% and 89%, SeaBank achieved 98% and 97%, Jago achieved 95% and 92%, and Jenius achieved 92% and 86%. For the Naive Bayes Classifier, the accuracy of training and testing data for Ovo achieved 82% and 77%, Dana achieved 84% and 81%, SeaBank achieved 95% and 93%, Jago achieved 86% and 82%, and Jenius achieved 86% and 80%. These results indicate that Support Vector Machine and Naive Bayes Classifier are highly accurate in sentiment analysis of e-Wallet and Digital Banking application reviews in Indonesia.
Item Type: | Thesis (Skripsi) | ||||||||
---|---|---|---|---|---|---|---|---|---|
Contributors: |
|
||||||||
Subjects: | 000 Karya Umum > 003 Sistem-sistem 000 Karya Umum |
||||||||
Divisions: | Fakultas Sains dan Teknologi > Sistem Informasi | ||||||||
Depositing User: | fsains - | ||||||||
Date Deposited: | 16 Jul 2024 01:46 | ||||||||
Last Modified: | 16 Jul 2024 01:46 | ||||||||
URI: | http://repository.uin-suska.ac.id/id/eprint/81969 |
Actions (login required)
![]() |
View Item |