PREDIKSI PENYAKIT PARU-PARU MENGGUNAKAN ALGORITMA NAÏVE BAYES DAN ADABOOST

Abstract

The lungs are the human respiratory system which plays an important role in meeting the body's oxygen needs. Apart from that, the lungs have a function as a place to exchange oxygen from the air with carbon dioxide from the blood. In some conditions, the lungs can experience problems which have a negative impact on the performance of the respiratory system. If the lungs do not function properly it will cause disease. The dataset used uses a dataset from Kaggle a  total of 30,000 data using rapid miner tools. The only attributes used are age, gender, work, household, activity of staying up late, sports activity, insurance, congenital disease. The methods used in this research are only Adaboost and Naïve Bayes. Based on the research that has been carried out, it can be concluded that research showing the use of the Adaboost and Naïve Bayes algorithms in predicting lung disease produces a better level of accuracy compared to using only Naïve Bayes. The analysis results show that the accuracy level of this research reached 94.66%, with a precision of 90.71% and a recall of 100.00%. In this experiment, using a combination of Naïve Bayes and adaboost succeeded in increasing the accuracy rate by 7.44%.