The Efficiency of Data Mining Technique for the Prognosis of Cerebrovascular Disease

Abstract

The purpose of this research is to compare the efficiency of data mining techniques to find the suitable model for prognosis of cerebrovascular disease. The 5110 patients with cerebrovascular disease was used in this research. The data was collected from Clínico San Carlos Hospital, Spain via www.kaggle.com. The data was analyzed in 4 different techniques of data mining consisting of (1) Naïve Bayes, (2) Deep Learning, (3) Decision Tree, and (4) Random Forest to simulate cerebrovascular disease prediction. The efficiency of data classification was compared in 3 different criteria which are accuracy, f-measure, and sensitivity to determine the most appropriate simulation for prognosis. The result suggested that the Deep Learning technique was the most appropriate technique to simulate prognosis of cerebrovascular disease with 95.11% of accuracy, 97.49% of f-measure, and 99.98% of sensitivity.

Keywords

• efficiency comparison
• data mining technique
• cerebrovascular disease