원문정보
초록
영어
Diabetes is a condition in which the amount of sugar in the blood is higher than normal. Classification systems have been widely used in the medical domain to explore patient’s data and extract a predictive model or set of rules. The prime objective of this research work is to facilitate a better diagnosis (classification) of diabetes disease. There are already several methodologies which have been implemented on classification for the diabetes disease. The proposed methodology implemented work in 2 stages: (a) In the first stage Genetic Algorithm (GA) has been used as a feature selection method on Pima Indian Diabetes Dataset (PIDD) (b) In the second stage, J48graft Decision Tree (J48graft DT) has been used for the classification and prediction on the selected feature. Early diagnosis of any disease with less cost is preferable. Diabetes is also one of such diseases. GA is noted to reduce not only the storage capacity, cost and computation time of the diagnostic process, but the proposed approach also improved the ROC of classification. The experimental results obtained classification accuracy (74.7826%) and ROC (0.786) show that GA and J48graft DT can be successfully used for the diagnosing of diabetes disease.
목차
1. Introduction
2. Related Work
3. Proposed Methodology
3.1. Proposed Algorithm
4. Results and Discussion
5. Conclusion and Future Work
References