PENINGKATAN AKURASI KLASIFIKASI ALGORITMA NAÏVE BAYES PADA PASIEN PENYAKIT JANTUNG MENGGUNAKAN METODE PARTICLE SWARM OPTIMIZATION (PSO) DI IHC RUMAH SAKIT PERTAMINA

The Particle Swarm Optimization (PSO) method is used in this study to improve the accuracy of using the Naive Bayes algorithm in classifying the life expectancy of heart disease patients. Heart disease data for 300 samples from Kaggle are used in this study. The study compares the classification accuracy before and after applying PSO with the Naive Bayes algorithm. In the three cases examined, the application of PSO proved successful in improving the classification accuracy. From 0.78 to 0.83, the accuracy increased by 5% in the first scenario. From 0.81 to 0.86, the accuracy also increased by 5% in the second scenario. Meanwhile, from 0.8 to 0.83 in the third example, the accuracy increased by 3%. Overall, PSO was able to improve the performance of the Naive Bayes algorithm, resulting in an increase in accuracy of 5%, 5%, and 3% in each case. In the context of heart disease data, these findings suggest that PSO can be used as an effective method to improve the accuracy of predictive models.