Sentiment Analysis of Health Insurance Application Reviews Using Naïve Bayes, Support Vector Machine, and SMOTE
Tạp chí Khoa học Lạc Hồng · 25.06.2026 · 1 (26) · 30-36
DOI: 10.61591/jslhu.26.1087 ↗Kurz gefasst
Fachartikel in der Zeitschrift Tạp chí Khoa học Lạc Hồng, online erschienen am 25.06.2026. Verfasst von Khau Van Bich und 2 weiteren Autor:innen.
Abstract (Originalsprache)
This study explores factors shaping user experience in mobile health insurance application by analyzing user reviews through sentiment classification. Two models such as Naïve Bayes (NB) and Support Vector Machine (SVM) were examined, with NB yielding higher precision and F1-score. Data preparation proved influential: applying the Synthetic Minority Oversampling Technique (SMOTE) improved average accuracy by 7.51%, slang normalization added a modest 0.25%, while including neutral sentiment reduced accuracy by 9.67%. The selected configuration, NB combined with SMOTE and slang replacement, achieved 93.53% accuracy, 93.65% precision, 93.53% recall, and an F1-score of 93.47%. Examination of reviews further shows that discussions about security, ease of use, and timeliness are dominated by negative sentiment.
Zitieren
APA
Bich, K. V., Tien, V. M., & Duat, T. H. (2026). Sentiment Analysis of Health Insurance Application Reviews Using Naïve Bayes, Support Vector Machine, and SMOTE. Tạp chí Khoa học Lạc Hồng, 1(26), 30-36. https://doi.org/10.61591/jslhu.26.1087
Vancouver
Bich KV, Tien VM, Duat TH. Sentiment Analysis of Health Insurance Application Reviews Using Naïve Bayes, Support Vector Machine, and SMOTE. Tạp chí Khoa học Lạc Hồng. 2026;1(26):30-36. doi:10.61591/jslhu.26.1087
BibTeX
@article{Bich2026,
author = {Bich, Khau Van and Tien, Vo Minh and Duat, Tran Huu},
title = {Sentiment Analysis of Health Insurance Application Reviews Using Naïve Bayes, Support Vector Machine, and SMOTE},
journal = {Tạp chí Khoa học Lạc Hồng},
year = {2026},
volume = {1},
number = {26},
pages = {30--36},
doi = {10.61591/jslhu.26.1087}
}
EUVIMED