PENGARUH KERNEL PADA SVM TERHADAP KLASIFIKASI SENTIMEN TWEET BAHASA INDONESIA

  • Titin Rahayu Universitas Tanjungpura
  • Arif Bijaksana Putra Negara Universitas Tanjungpura
  • Enda Esyudha Pratama Universitas Tanjungpura
Keywords: Sentiment Analysis, Twitter, Support Vector Machine, SMOTE, Grid Search, Kernel.

Abstract

Sentiment analysis of Indonesian-language Twitter data presents challenges due to the short, unstructured nature of tweets, as well as the frequent use of informal language, abbreviations, and non-standard words. This study aims to analyze the effect of Support Vector Machine (SVM) kernel selection, Synthetic Minority Over-sampling Technique (SMOTE), and hyperparameter tuning using Grid Search on the performance of sentiment classification of Indonesian-language tweets. The dataset consisted of 10,820 tweets classified into three sentiment categories: positive, negative, and neutral. The research stages included text preprocessing, TF-IDF feature extraction, data splitting, SMOTE implementation, SVM modeling using Linear, Radial Basis Function (RBF), Sigmoid, and Polynomial kernels, and model evaluation using accuracy, precision, recall, and F1-score. A total of 16 scenarios were tested based on combinations of kernels, SMOTE, and Grid Search. The results showed that the RBF kernel achieved the best performance. The combination of RBF kernel, SMOTE, and Grid Search obtained an accuracy of 75.47%, precision of 75.95%, recall of 75.47%, and F1-score of 75.54%. Therefore, the combination of RBF kernel, SMOTE, and Grid Search was the best configuration for sentiment classification of Indonesian-language tweets in this dataset.

References

Aulia, T. M. P. (2021). Perbandingan kernel Support Vector Machine (SVM) dalam penerapan analisis sentimen vaksinisasi COVID-19. SINTECH (Science and Information Technology) Journal, 4(2), 139–145. https://doi.org/10.31598/sintechjournal.v4i2.762
Darwis, M., Pranoto, G. T., Wicaksana, Y. E., & Yaddarabullah, Y. (2020). Implementation of TF-IDF algorithm and K-mean clustering method to predict words or topics on Twitter. JISA (Jurnal Informatika dan Sains), 3(2), 49–55. https://doi.org/10.31326/jisa.v3i2.831
Ferdiana, R., Jatmiko, F., Purwanti, D. D., Tri Ayu, A. O., & Dicka, R. (2019). Dataset Indonesia untuk analisis sentimen. Proceedings of the 2019 International Conference on Information and Communication Technology.
Jumeilah, F. S. (2017). Penerapan Support Vector Machine (SVM) untuk pengkategorian penelitian. Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), 1(1), 19–25. https://doi.org/10.29207/resti.v1i1.11
Khairunnisa, S., Adiwijaya, A., & Faraby, S. Al. (2021). Pengaruh text preprocessing terhadap analisis sentimen komentar masyarakat pada media sosial Twitter (Studi kasus pandemi COVID-19). Jurnal Media Informatika Budidarma, 5(2), 406. https://doi.org/10.30865/mib.v5i2.2835
Meidianingsih, Q., & Meganingtyas, D. E. W. (2022). Analisis perbandingan performa metode ensemble dalam menangani imbalanced multi-class classification. Jurnal Aplikasi Statistika & Komputasi Statistik, 14(2), 13–21. https://doi.org/10.34123/jurnalasks.v14i2.335
Passonneau, R. (2011). Sentiment analysis of Twitter data. 30–38.
Pratama, E. E., & Trilaksono, B. R. (2015). Klasifikasi topik keluhan pelanggan berdasarkan tweet dengan menggunakan penggabungan feature hasil ekstraksi pada metode Support Vector Machine (SVM). Jurnal Edukasi dan Penelitian Informatika (JEPIN), 1(2). https://doi.org/10.26418/jp.v1i2.11023
Rozi, I. F., Ardiansyah, R., & Rebeka, N. (2019). Penerapan normalisasi kata tidak baku menggunakan Levenshtein Distance pada analisa sentimen layanan PT. KAI di Twitter. Seminar Informatika Aplikatif, 106–112.
Sutoyo, E., & Fadlurrahman, M. A. (2020). Penerapan SMOTE untuk mengatasi imbalance class dalam klasifikasi Television Advertisement Performance Rating menggunakan Artificial Neural Network. Jurnal Edukasi dan Penelitian Informatika (JEPIN), 6(3), 379. https://doi.org/10.26418/jp.v6i3.42896
Utomo, M. S. (2013). Implementasi Stemmer Tala pada aplikasi berbasis web. Jurnal Teknologi Informasi DINAMIK, 18(1), 41–45.
Published
2026-09-25
How to Cite
Titin Rahayu, Arif Bijaksana Putra Negara, & Enda Esyudha Pratama. (2026). PENGARUH KERNEL PADA SVM TERHADAP KLASIFIKASI SENTIMEN TWEET BAHASA INDONESIA. EDU RESEARCH, 7(3), 1414-1423. https://doi.org/10.47827/jer.v7i3.3801
Section
Articles