ANALYSIS OF CUSTOMER SATISFACTION IN PURCHASING SKINCARE IN THE TOKOPEDIA APPLICATION USING TEXT CLASSIFICATION AND TOPIC MODELING
Abstract
This study aims to analyze customer satisfaction in purchasing skincare products on the
Tokopedia platform using sentiment classification and topic modeling approaches. Amid the
growing trend of online shopping, customer reviews as a form of User-Generated Content (UGC)
have become a crucial source of information that directly reflects consumers’ experiences and
perceptions. The study uses 14,716 reviews from five leading local skincare brands, analyzed using
the IndoBERT model for sentiment classification and BERTopic to identify the main themes across
several e-satisfaction dimensions, namely: Delivery, Product Quality, Offers & Discount, and
Customer Support. The analysis shows that the Delivery dimension most frequently receives
negative sentiment, while Product Quality dominates positive reviews. Topic modeling reveals key
issues discussed by users, such as delayed shipping, packaging conditions, product quality, and
promotional programs, which reflect both the strengths and weaknesses of the services. These
findings provide strategic insights for e-commerce platforms and skincare brands to enhance
logistics, maintain product quality, refine marketing strategies, and improve the overall customer
experience.
Tokopedia, Customer Satisfaction, Skincare, Sentiment Analysis, BERTopic, ESatisfaction
References
Alzahrani, M., Aldhyani, T. H. H., Alsubari, S. N., & Althobaiti, M. M. (2022). Developing an
intelligent system with deep learning algorithms for sentiment analysis of e-commerce
product reviews. Computational Intelligence and Neuroscience, 2022(2), 1–10.
https://doi.org/10.1155/2022/3840071
Candiwan, C., & Wibisono, C. (2021). Analysis of the Influence of Website Quality to Customer’s
Loyalty on e-commerce. International Journal of Electronic Commerce Studies, 12(1),
https://doi.org/10.7903/ijecs.1892
Firdaus, R. Z., Hadi Wijoyo, S., & Purnomo, W. (2025). Analisis Sentimen Berbasis Aspek Ulasan
Pengguna Aplikasi Alfagift Menggunakan Metode Random Forest dan Pemodelan Topik
Latent Dirichlet Allocation (Vol. 9, Issue 2). http://j-ptiik.ub.ac.id
Jiang, X. (2024). Statistical and sentiment analysis based on comments of skin care products. In
International Conference on Signal Processing and Machine Learning (SPML) (pp.
–128). IEEE. https://doi.org/10.1109/SPML59876.2024.10325127
Kumar, A., Chakraborty, S., & Bala, P. K. (2023a). Text mining approach to explore determinants
of grocery mobile app satisfaction using online customer reviews. Journal of Retailing
and Consumer Services, 73. https://doi.org/10.1016/j.jretconser.2023.103363
Mahfudiyah, N., & Alamsyah, A. (2022). Analisis Persepsi Konsumen Terhadap Kualitas Layanan
Gojek Menggunakan Sentiment Analysis Dan Topic Modeling Berdasarkan Deep
Learning IndoBERT Consumer Perception Analysis Of Gojek Service Quality Using
Sentiment Analysis And Topic Modeling Based On Deep Learning IndoBERT
Prasetio, A., Rahman, D. A., Sary, F. P., Pasaribu, R. D., & Sutjipto, M. R. (2022). The role of
Instagram social media marketing activities and brand equity towards airlines customer
response. International Journal of Data and Network Science, 6(4), 1195–
https://doi.org/10.5267/j.ijdns.2022.6.014
Rachmawati, I., & Perdani. (2020). Influence of the Use of Brands in a Foreign Language
(English) on the Purchase Decision Process of Cosmetic Products Made in Indonesia.
In Pertanika J. Soc. Sci. & Hum. Universiti Putra Malaysia
Press. http://www.pertanika.upm.edu.my/
Sekaran, U., & Bougie, R. (2016). Research Methods for Business: A Skill-Building Approach
(Seventh Edition). John Wiley & Sons Ltd.
Wu, C. H. J., & Chang, Y. Y. (2020). Exploring customer sentiment regarding online retail
services: A topic-based approach. Journal of Retailing and Consumer Services, 54, 102001.
