Sentiment analysis of consultation answers on Alodokter using Support Vector Machine (SVM)
Keywords:
Support Vector Machine, Sentiment Analysis, Alodokter, Health InformaticsAbstract
In the medical field, fast and accurate access to information is crucial for diagnosis, treatment, and research. Medical information retrieval systems play an essential role in facilitating this access. Despite the abundance of online medical resources, users often face challenges in processing and interpreting information efficiently. Question Answering (QA) systems aim to provide accurate and relevant responses to user queries, making them integral to platforms such as Alodokter, one of the most popular health websites in Indonesia. Alodokter offers vast medical content and allows direct interactions between users and certified medical professionals, making it a rich source of reliable and contextually relevant data. This study explores the implementation of the Support Vector Machine (SVM) method to classify and analyze sentiment in responses found on Alodokter. SVM is a supervised machine learning algorithm known for its high performance in classification tasks, particularly with non-linearly separable data. Its strong generalization capabilities make it well-suited for complex QA data involving diverse linguistic structures and varying medical contexts. Using Alodokter data, this research evaluates SVM performance across different training-testing ratios to identify the most effective configuration. The experimental results demonstrate that the best classification performance was achieved with a 60:40 training-testing ratio, yielding an accuracy of 70%. At this ratio, negative questions achieved 68% precision, 78% recall, and an F1-score of 73%, while positive questions yielded 74% precision, 62% recall, and F1-score of 67%. Conversely, the 50:50 ratio resulted in the lowest accuracy of 59%, with notable imbalances in recall values, particularly for negative questions. In conclusion, SVM proves to be an effective tool for sentiment-based QA analysis on Alodokter, offering valuable insights to improve online health services and enhance user experience in digital healthcare platforms.
References
Abdillah, A.F., Putra, C.B.P., Apriantoni, A., Juanita, S. dan Purwitasari, D., (2022). Ensemble-based Methods for Multi-label Classification on Biomedical Question-Answer Data. Journal of Information Systems Engineering and Business Intelligence, 8(1), pp.42–50
Aruda, S. A. Q. (2022). Klasifikasi Pertanyaan Berbahasa Indonesia Menggunakan Algoritma Support Vector Machine. 14(2), 44–52.
Fikri, M. I., Sabrila, T. S., & Azhar, Y. (2020). Perbandingan Metode Naïve Bayes dan Support Vector Machine pada Analisis Sentimen Twitter. Smatika Jurnal, 10(02), 71–76. https://doi.org/10.32664/smatika.v10i02.455
Hanami, A. (2023). Term Frequency Inverse Document Frequency (TF-IDF). Visitor Analytics, December. https://www.visitor-analytics.io/es/glosario/t/term-frequency-inverse-document-frequency-tf-idf
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
Lestari, T. P. (2022). Analisis Text Mining pada Sosial Media Twitter Menggunakan Metode Support Vector Machine (SVM) dan Social Network Analysis (SNA). Jurnal Informatika Ekonomi Bisnis, 4(3), 65–71. https://doi.org/10.37034/infeb.v4i3.146
Murti Ali Lingga, E. D. (2019). Alodokter, Konsultasi Kesehatan Via Online. https://money.kompas.com/read/2019/04/25/204700726/alodokter-konsultasi-kesehatan-via-online?page=all
Rajpurkar, P., Zhang, J., Lopyrev, K., & Liang, P. (2016). SQuad: 100,000+ questions for machine comprehension of text. EMNLP 2016 - Conference on Empirical Methods in Natural Language Processing, Proceedings, ii, 2383–2392. https://doi.org/10.18653/v1/d16-1264
Rama, P., Putra, B., & Perdana, R. S. (2023). Klasifikasi Judul Berita Online menggunakan Metode Support Vector Machine ( SVM ) dengan Seleksi Fitur Chi-square. Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komputer, 7(5), 2132–2141.
Siringoringo, R., & Jamaludin, J. (2019). Text Mining dan Klasterisasi Sentimen Pada Ulasan Produk Toko Online. Jurnal Teknologi Dan Ilmu Komputer Prima (JUTIKOMP), 2(1), 41–48. https://doi.org/10.34012/jutikomp.v2i1.456.
Satria, G. J., Adikara, P. P., & Wihandika, R. C. (2022). Klasifikasi Pertanyaan COVID-19 Bahasa Indonesia menggunakan Naive Bayes. Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komputer, 6(1), 148–153. https://j-ptiik.ub.ac.id/index.php/j-ptiik/article/view/10382
Sudin, S., Junaedi, H., & Santosa, J. (2019). Analisis Jenis Pertanyaan Berbahasa Indonesia pada Question and Answering System Menggunakan Metode Support Vector Machine(SVM). Dintek, 12(1), 72–80.
Winiarti, S., Widayanti, D., & Ahdiani. (2022). Klasifikasi Jenis Buku Berdasarkan Cover dan Judul Buku Menggunakan Metode Support Vector Machine dan Cosine Similarity. Sainteks, 19(1), 53. https://doi.org/10.30595/sainteks.v19i1.13423
Yasni, L., Subroto, I. M. I., & Haviana, S. F. C. (2018). Implementasi Cosine Similarity Matching Dalam Penentuan Dosen Pembimbing Tugas Akhir. Transmisi, 20(1), 22. https://doi.org/10.14710/transmisi.20.1.22-28
Zhang, X., Wu, J., He, Z., Liu, X., & Su, Y. (2018). Medical exam question answering with large-scale reading comprehension. 32nd AAAI Conference on Artificial Intelligence, AAAI 2018, 5706–5713. https://doi.org/10.1609/aaai.v32i1.11970
