Analisis Sentimen Ulasan Aplikasi TikTok Menggunakan Convolutional Neural Network

Kelvin Dwi Pranata Sembiring; Agam Setiawan; Harun Rosyid

doi:10.31004/riggs.v4i4.4860

Authors

Kelvin Dwi Pranata Sembiring Universitas Negeri Surabaya
Agam Setiawan Universitas Negeri Surabaya
Harun Rosyid Universitas Negeri Surabaya

DOI:

https://doi.org/10.31004/riggs.v4i4.4860

Keywords:

Analisis Sentimen, TikTok, Deep Learning, CNN, Machine Learning, NLP

Abstract

Analisis sentimen terhadap ulasan pengguna merupakan instrumen krusial dalam memahami persepsi serta pengalaman pengguna terhadap aplikasi mobile seperti TikTok. Di tengah pesatnya pertumbuhan platform digital, pemrosesan umpan balik secara otomatis menjadi kebutuhan mendesak bagi pengembang. Penelitian ini bertujuan untuk menganalisis sentimen ulasan aplikasi TikTok dengan membandingkan efektivitas metode machine learning dan deep learning, yaitu Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Support Vector Machine (SVM), dan Logistic Regression (LR). Data penelitian diperoleh melalui teknik scraping pada Google Play Store periode Oktober 2025, menghasilkan 19.569 ulasan mentah yang kemudian diproses menjadi 11.368 data bersih melalui tahapan preprocessing. Tahapan tersebut meliputi pembersihan teks, stopword removal, tokenisasi, serta pelabelan sentimen dan kategori dengan rasio pembagian data 80:20. Evaluasi performa model dilakukan secara komprehensif menggunakan metrik Accuracy, Macro-F1, dan Weighted-F1. Hasil eksperimen menunjukkan bahwa model CNN mencapai performa terbaik dengan nilai Accuracy 0,9895, Macro-F1 0,9407, dan Weighted-F1 0,9897, secara konsisten mengungguli LSTM, SVM, dan Logistic Regression. Skor tinggi ini membuktikan bahwa arsitektur CNN sangat efektif dalam mengekstraksi fitur lokal dan menangani karakteristik teks pendek yang informal pada ulasan media sosial. Temuan ini memberikan kontribusi signifikan bagi peneliti NLP dalam mengoptimalkan sistem klasifikasi sentimen, serta memberikan wawasan strategis bagi pengembang aplikasi untuk meningkatkan kualitas layanan berdasarkan sentimen pengguna secara akurat.

Downloads

Download data is not yet available.

References

M. A. H. Ahmad Asmawi, P. Isawasan, and L. Shamugam, “Exploring sentiment trends in TikTok comments using GPT for influencer content strategy,” e-Academia J., 2025, [Online]. Available: https://ir.uitm.edu.my/id/eprint/118867/1/118867.pdf

R. Kansal and C. Diwaker, “Efficiency determination of various machine learning techniques for sentiment analysis on social media platforms,” Eng. Technol. Appl. Sci. Res., vol. 15, no. 1, pp. 1234–1240, 2025, [Online]. Available: https://etasr.com/index.php/ETASR/article/download/11158/5339

D. D. N. Cahyo, R. Handayani, and V. B. Lestari, “Sentiment analysis of public opinion on online gambling through social media using convolutional neural network,” J. Informatics Telecommun. Eng., vol. 9, no. 2, pp. 115–124, 2025, [Online]. Available: https://ojs.uma.ac.id/index.php/jite/article/download/15024/6502

S. Kurniawan, A. S. Pramayoga, and Y. F. Ashari, “Sentiment analysis in coastal tourism: A systematic literature review,” Eng. Technol. Appl. Sci. Res., vol. 15, no. 3, pp. 2311–2321, 2025, [Online]. Available: https://etasr.com/index.php/ETASR/article/download/14644/5986

M. Dixit, D. Sharma, and V. Singh, “Video sentiment analysis on social media using an advanced VADER technique,” J. Data Sci. Anal., 2025, [Online]. Available: https://doi.org/10.1007/s41060-025-00849-1

L. Zhiming, H. Huijian, and L. Zongwei, “Predicting the Shanghai Composite Index using Chinese TikTok self-media data and machine learning model in China,” Discret. Dyn. Nat. Soc., vol. 2024, pp. 1–12, 2024, [Online]. Available: https://doi.org/10.1155/ddns/7201831

Q. Shen, S. Han, Y. Han, and X. Chen, “User review analysis of dating apps based on text mining,” PLoS One, vol. 18, no. 4, p. e0283896, 2023.

J. Dąbrowski, E. Letier, A. Perini, and A. Susi, “Analysing app reviews for software engineering: A systematic literature review,” Empir. Softw. Eng., vol. 27, no. 2, p. 43, 2022.

M. Das and P. J. A. Alphonse, “A comparative study on TF-IDF feature weighting method and its analysis using unstructured dataset.” 2023. [Online]. Available: arxiv:2308.04037

A. H. Dani, E. Y. Puspaningrum, and R. Mumpuni, “Studi performa TF-IDF dan Word2Vec pada analisis sentimen cyberbullying,” Router J. Tek. Inform. dan Terap., vol. 2, no. 2, pp. 94–106, 2024.

V. B. Lestari and C. A. Hutagalung, “Evaluation of TF-IDF extraction techniques in sentiment analysis of Indonesian-language marketplaces using SVM, logistic regression, and naive Bayes,” J-KOMA J. Comput. Sci. Appl., vol. 8, no. 1, pp. 36–44, 2025.

A. Hermawan, I. Jowensen, and J. Junaedi, “Implementasi text-mining untuk analisis sentimen pada Twitter dengan algoritma support vector machine,” JST (Jurnal Sains dan Teknol., vol. 12, no. 1, pp. 129–137, 2023.

M. Prasetya, M. Wulandari, and S. A. Nikmah, “Implementasi NLP (Natural Language Processing) Dasar pada Analisis Sentiment Review Spotify BT - Seminar Nasional Teknologi & Sains,” 2024, pp. 145–153.

J. Opitz, “A closer look at classification evaluation metrics and a critical reflection of common evaluation practice,” Trans. Assoc. Comput. Linguist., vol. 12, pp. 820–836, 2024.

W. Zhao, R. Singh, T. Joshi, A. Sudjianto, and V. N. Nair, “Self-interpretable convolutional neural networks for text classification.” 2021. [Online]. Available: arxiv:2105.08589

S. Soni, S. S. Chouhan, and S. S. Rathore, “TextConvoNet: A convolutional neural network-based architecture for text classification,” Appl. Intell., vol. 53, no. 11, pp. 14249–14268, 2023.

H. Cao, “Recent advances in text embedding: A comprehensive review of top-performing methods on the MTEB benchmark.” 2024. [Online]. Available: arxiv:2406.01607