FRAYUDI, NPM 2107100021 (2024) COMPARATIVE ANALYSIS OF MACHINE LEARNING ALGORITHMS IN SENTIMENT ANALYSIS OF E-COMMERCE APPLICATION USER REVIEWS. Tugas_Akhir (Artikel) International Journal of Science, Technology & Management, 5 (4). ISSN 2722 - 4015 (e-ISSN)
![[img]](http://repository.ulb.ac.id/style/images/fileicons/text.png) |
Text
COVER.pdf Download (1MB) |
|
Text
ARTIKEL.pdf Download (265kB) |
Abstract
Shopee, as one of the leading e-commerce platforms, receives thousands of user reviews every day. These reviews contain valuable information that can help improve the quality of products and services. However, analyzing reviews manually is a very time-consuming task and is prone to human error. Therefore, the analysis of user review sentiment requires an automated solution. Machine learning algorithms in user review sentiment analysis can help solve this problem more efficiently. Algorithms such as SVM and Naïve Bayes can be used to classify reviews as positive or negative, providing actionable insights for management. This study aims to compare the performance of two machine learning algorithms, namely SVM and Naïve Bayes, in analyzing the sentiment of Shopee app user reviews. We collected and cleaned Shopee app user review data for the model. Next, we divided the data into a training set and a test set. Both algorithms, SVM and Naïve Bayes, were trained using the training set and evaluated using the test set. The results showed that Naïve Bayes had a precision of 81% and SVM had a precision of 80%. In terms of recall, SVM is superior with a value of 80% compared to Naïve Bayes, which has a value of 79%. Both algorithms have the same F1-score, which is 80%. In terms of accuracy, SVM is slightly superior with a value of 80% compared to Naïve Bayes, which has a value of 79%. Based on the evaluation results, SVM shows a slightly better performance than Naive Bayes in the sentiment analysis of Shopee application user reviews. Although Naive Bayes is superior in precision, SVM has higher recall and accuracy, making it the best algorithm overall in this study. E commerce companies like Shopee can use this study's insights into the performance of two machine learning algorithms in sentiment analysis of user reviews to enhance the quality of their products and services. In addition, this study also provides guidance for researchers and practitioners in choosing the right algorithm for sentiment analysis based on their specific needs. Thus, this study not only helps in understanding the performance of SVM and Naïve Bayes algorithms but also provides a solid foundation for the practical implementation of sentiment analysis in the context of e-commerce. Keywords : E-commerce, Machine Learning, Naïve Bayes, Sentiment Analysis, Shopee, SVM.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | E-commerce, Machine Learning, Naïve Bayes, Sentiment Analysis, Shopee, SVM. |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Sains Dan Teknologi > Informatika Komputer |
| Depositing User: | Unnamed user with email repository@ulb.ac.id |
| Date Deposited: | 09 Oct 2024 07:34 |
| Last Modified: | 09 Oct 2024 07:34 |
| URI: | http://repository.ulb.ac.id/id/eprint/1169 |
Actions (login required)
 |
View Item |
