Optimization the Naive Bayes algorithm using particle swarm optimization feature selection and bagging techniques for detection of Alzheimer’s disease

• Published in: AIP conference proceedings Vol. 2714; no. 1
• Main Authors: Saputra, Rizal Amegia, Puspitasari, Diah, Wahyudi, Mochamad, Ramdhani, Lis Saumi, Ramanda, Kresna
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Melville American Institute of Physics 09.05.2023
• Subjects: Accuracy; Algorithms; Alzheimer's disease; Bagging; Data mining; Feature selection; Optimization; Particle swarm optimization
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0128553

Alzheimer’s is a deadly disease it can cause dementia in sufferers. It is necessary for early detection in the treatment of this disease. Many studies have discussed Alzheimer’s disease with data mining techniques, but the most accurate method is unknown. This paper proposed a Naive Bayes algorithm with Particle Swarm Optimization (PSO) selection feature and bagging for optimize unbalanced data. The results of the experiment with 10-fold cross validation, the first test using naive bayes algorithm obtained an accuracy value of 93.75%, with a AUC value of 0.966. Furthermore, the test used with PSO feature selection and bagging technique, and the accuracy value obtained by 98.21% with a AUC value of 0.989. The results of this test can be concluded that the testing of PSO feature selection and bagging techniques, the accuracy value obtained has increased significantly, this proves that the optimization of algorithms with PSO feature selection and bagging techniques has excellent classification.