Sentiment Analysis of Facial Wash Sombong Product Reviews Using the Naive Bayes Algorithm
Article Sidebar
Main Article Content
Abstract
This study aims to analyze public sentiment toward the Facial Wash Sombong product, a beauty product that went viral on social media, using Naïve Bayes Classification (NBC). The data were obtained from 5,000 user comments on the TikTok platform, collected through web crawling techniques. The data preprocessing stages included cleansing, case folding, tokenizing, filtering, and stemming to prepare the textual data. The results of the sentiment distribution analysis indicate a dominance of negative opinions at 61.4% (3,070 comments), while positive sentiment accounts for 38.6% (1,930 comments). Keyword analysis reveals that negative sentiment is mainly triggered by issues related to side effects, such as acne and skin irritation. The NBC model was evaluated using an 80:20 data split and achieved strong performance, with an accuracy of 85.62%, sensitivity of 88.40%, and specificity of 76.90%. The effective performance of the model validates the use of NBC for sentiment analysis and provides strategic recommendations for producers to address public criticism.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1] E. Apriani, I. F. Hanif, and F. Oktavianalisti, “Sentiment Analysis of Using TikTok as a Learning Media Using the Naïve Bayes Classifiers Algorithm Analisis Sentimen Penggunaan TikTok Sebagai Media Pembelajaran Menggunakan Algoritma Naïve Bayes Classifier,” vol. 4, no. July, pp. 1160–1168, 2024.
[2] D. Metode, S. Vector, and M. Svm, “Inti nusa mandiri,” vol. 19, no. 2, pp. 325–332, 2025.
[3] D. Amalia, M. H. Totohendarto, and S. Alam, “Analisis Sentimen Produk Populer Moisturizer Pada Female Daily Menggunakan Metode Naive Bayes,” vol. 8, no. 2, pp. 108–121, 2024.
[4] S. Rahmawati, “Implementasi Algoritma Bert Untuk Analisis Sentimen Ulasan Pengguna Aplikasi Pedulilindungi,” pp. 6–19, 2023.
[5] T. Astuti and Y. Astuti, “Analisis Sentimen Review Produk Skincare Dengan Naïve Bayes Classifier Berbasis Particle Swarm Optimization ( PSO ),” vol. 6, pp. 1806–1815, 2022, doi: 10.30865/mib.v6i4.4119.
[6] B. Najibah, A. Ratri, and Y. A. Sari, “Analisis Sentimen Review Produk Kecantikan menggunakan Metode Naïve Bayes,” vol. 5, no. 12, pp. 5635–5641, 2021.
[7] M. Z. Maharani and U. Padjadjaran, “Analisis Sentimen Positif Terhadap Avoskin sebagai Eco Friendly Brand di Media Sosial X dan TikTok,” no. 3, 2024.
[8] R. Sistem, M. Lestandy, A. Abdurrahim, and L. Syafa, “Analisis Sentimen Tweet Vaksin COVID-19 Menggunakan Recurrent,” vol. 5, no. 10, pp. 802–808, 2021.
[9] M. F. Alhafizh, “Analisis Sentimen Dengan Naive Bayes Classifier Menggunakan Tf-Idf Dan N-Gram,” vol. 10120025, 2024.
[10] U. Analisis and S. Berbasis, “Jurnal Indonesia : Manajemen Informatika Dan Komunikasi Penerapan Algoritma Tf-Idf Dan Naïve Bayes Jurnal Indonesia : Manajemen Informatika Dan Komunikasi,” vol. 4, no. 3, pp. 1822–1834, 2023.
[11] D. Mustikananda, D. E. Ratnawati, and B. Rahayudi, “Perbandingan Algoritma Naïve Bayes dan Support Vector Machine untuk Analisis Sentimen terhadap Review Produk Aster Kosmetik Malang Marketplace Shopee,” vol. 6, no. 7, pp. 3137–3144, 2022.
[12] P. Studi, T. Informatika, J. T. Informatika, F. I. Komputer, and U. Brawijaya, “Analisis Sentimen Ulasan Produk Kecantikan Mengunakan Metode Bm25 Dan Improved K-Nearest Neighbor Dengan Seleksi Fitur Chi-Square,” 2020.
[13] S. Nurjanah and Y. H. Apidana, “Analisis Sentimen TikTok untuk Mengevaluasi Reputasi Merek Pasca Kasus Overclaim : Studi pada Daviena Skincare,” vol. 4, no. 2, 2025.
[14] F. Tf-idf and T. Razaq, “Analisis Sentimen Review Film Menggunakan Naive Bayes Classifier Dengan,” vol. 10, no. 2, pp. 1698–1712, 2023.
[15] N. F. Hilmi, “Analisis Sentimen Terhadap Aplikasi Tiktok Dari Ulasan Pada Google Playstore Menggunakan Metode Naïve Bayes,” vol. 14, no. 1, pp. 146–156, 2024.