DOI:10.20894/IJCOA.
Periodicity: Bi Annual.
Impact Factor:
SJIF:5.079 & GIF:0.416
Submission:Any Time
Publisher: IIR Groups
Language: English
Review Process:
Double Blinded
The enormous growth of data in biomedical and healthcare communities need accurate analysis of medical data benefits in early disease detection, patient care and society services. However, the accuracy of analysis data will be condensed when the eminence of health care data is imperfect. In heath care community different regions exist with regional disease which would not be easily predicted with maximum of accuracy level. In this paper, we predict the diabetes disease and compare the algorithm which algorithm provide high performance, finally select the best algorithm to predict the diabetes disease at early stage. Machine learning algorithm can be applied for diabetes disease automating classification. This paper compares several Machine learning algorithms for classifying diabetes disease. Algorithms that involve Decision Tree, Naive Bayes, and KNN, SVM are proposed and assessed for this classification. These approaches have been tested with PIMA Indian Diabetes Dataset downloaded from UCI machine learning data repository. The performances of the algorithms have been compared in terms of Accuracy, Sensitivity, and Specificity with help of Sciklit-learn. Sciklit-learn are a free software machine learning library for the Python programming language. Finally comes with best suitable model for predict diabetes diseases.
