PERBANDINGAN KINERJA ALGORITMA NAIVE BAYES DAN C4.5 DALAM PREDIKSI PENYAKIT JANTUNG

Sri Wulandari; Kusrini Kusrini; Hanafi Hanafi

doi:10.46764/teknimedia.v6i2.284

Sri Wulandari Megiter Teknik Informatika Universitas Amikom Yogyakarta
Kusrini Kusrini Megiter Teknik Informatika Universitas Amikom Yogyakarta
Hanafi Hanafi Megiter Teknik Informatika Universitas Amikom Yogyakarta

DOI: https://doi.org/10.46764/teknimedia.v6i2.284

Keywords: Comparison, Naïve Bayes, C4.5, Confusion Matrix

Abstract

Information Technology is a data processing technology and is a variety of ways to produce high-quality information accurately and quickly, relevant to the needs of individuals and businesses. Strategic information about decision making. The development of information technology is one of the most important factors for the progress of time. There are several fields that are important for technological progress and affect the progress of the country, such as the education sector, the economic sector, the health sector, the government sector, and the socio-cultural sector. Basically, technology is developed to promote human work. Currently, technology is a great need for humanity. In fact, technology is used in all aspects of human life. Predicting heart disease accurately is essential to treat heart patients efficiently before a heart attack occurs. This goal can be achieved by using an optimal machine learning model with complete heart disease health data. Therefore, a comparison of the performance of the Naive Bayes algorithm and the C4.5 algorithm in predicting heart disease requires calculation so that the results obtained are more accurate. Before doing the calculation, it is necessary to check the feasibility of the data to be used, then the division of training and testing data. In the study, there were several scenarios for dividing training and testing data using a confusion matrix. This study resulted in a performance comparison of the Naïve Bayes and C4.5 algorithms in predicting heart disease, 6 experimental scenarios were carried out, each algorithm had 3 experiments with varying amounts of training data and testing data. The C4.5 algorithm performed 3 experimental scenarios, in the first experiment the Naïve Bayes algorithm, the first experiment 70:30 produced an accuracy of 83%. In the second experiment 80:20 produced an accuracy of 83%. In the third experiment 90:10 produced an accuracy of 85%. Then the C4.5 algorithm performed 3 experimental scenarios, in the first experiment 70:30 produced an accuracy of 98%. In the second experiment 80:20 produced an accuracy of 100% In the third experiment 90:10 produced an accuracy of 100%.

References

[1] Cecep Abdul Cholik, “Teknologi Informasi, ICT,” J. Fak. Tek., vol. 2, no. 2, pp. 39–46, 2021.
[2] A. Sepharni, I. E. Hendrawan, and C. Rozikin, “Klasifikasi Penyakit Jantung dengan Menggunakan Algoritma C4.5,” STRING (Satuan Tulisan Ris. dan Inov. Teknol., vol. 7, no. 2, p. 117, 2022, doi: 10.30998/string.v7i2.12012.
[3] T. A. Afifah, R. Martiansah, M. Alviyoni, and A. Arifin, “Perbandingan Algoritma Klasifikasi C4.5 Dan Naive Bayes untuk Memprediksi Gagal Jantung,” SENTIMAS Semin. Nas. Penelit. dan Pengabdi. Masy., pp. 78–85, 2023, [Online]. Available: https://www.kaggle.com/datasets/fedesoriano/heart-failure-prediction,
[4] C. L. Rohmat, R. Tito Aprilianto, F. Fathurrohman, and I. Iin, “Analisis Sentimen Pada Ulasan Produk Pakaian Zirdigo Dengan Menggunakan Algoritma Naive Bayes,” JATI (Jurnal Mhs. Tek. Inform., vol. 8, no. 1, pp. 668–674, 2024, doi: 10.36040/jati.v8i1.8697.
[5] N. Leniawati and S. Wijayanto, “Klasifikasi Desa Wisata di Provinsi Jawa Tengah Menggunakan Algoritma C4.5,” J. Sains Komput. Inform. (J-SAKTI, vol. 8, no. 1, pp. 171–184, 2024.
[6] F. Fatmawati and N. Narti, “Perbandingan Algoritma C4.5 dan Naive Bayes Dalam Klasifikasi Tingkat Kepuasan Mahasiswa Terhadap Pembelajaran Daring,” JTIM J. Teknol. Inf. dan Multimed., vol. 4, no. 1, pp. 1–12, 2022, doi: 10.35746/jtim.v4i1.196.
[7] D. Septiani, “Dan Naive Bayes Untuk Prediksi Penyakit Hepatitis,” J. Pilar Nusa Mandiri, vol. 13, no. 1, pp. 76–84, 2017, [Online]. Available: http://archive.ics.uci.edu/ml/.
[8] H. Syahrizal and M. S. Jailani, “Jenis-Jenis Penelitian Dalam Penelitian Kuantitatif dan Kualitatif,” J. QOSIM J. Pendidik. Sos. Hum., vol. 1, no. 1, pp. 13–23, 2023, doi: 10.61104/jq.v1i1.49.
[9] M. M. Ali, T. Hariyati, M. Y. Pratiwi, and S. Afifah, “Metodologi Penelitian Kuantitatif dan Penerapannya dalam Penelitian,” Educ. Journal.2022, vol. 2, no. 2, pp. 1–6, 2022.
[10] A. S. Setyanegara and E. Zuliarso, “Menerapkan Data Mining Dengan Algoritma Fp-Growth Pada Analisis Pola Pembelian Konsumen Pada Donat Bolong Semarang,” INTECOMS J. Inf. Technol. Comput. Sci., vol. 6, no. 2, pp. 866–872, 2023, doi: 10.31539/intecoms.v6i2.7106.
[11] M. Z. Haq, C. S. Octiva, A. Ayuliana, U. W. Nuryanto, and D. Suryadi, “Algoritma Naïve Bayes untuk Mengidentifikasi Hoaks di Media Sosial,” J. Minfo Polgan, vol. 13, no. 1, pp. 1079–1084, 2024, doi: 10.33395/jmp.v13i1.13937.
[12] F. R. Valerian et al., “Klasifikasi tingkat obesitas menggunakan metode gbm dan confusion matrix,” vol. 9, no. 2, pp. 2242–2249, 2025.