期刊名称：International Journal of Advances in Engineering and Management
电子版ISSN：2395-5252
出版年度：2021
卷号：3
期号：7
页码：1547-1562
DOI：10.35629/5252-030714141418
语种：English
出版社：IJAEM JOURNAL
摘要：In a recent survey it was discovered that the healthcare industry needs accurate data of increasing cases of diabetes mellitus. Today it is a big question mark for the whole world in our health care industry to predict and prevent diabetic mellitus, which is a chronic and fatal disease for a human being. Primarily it occurs in our body when the pancreas does not produce enough insulin or when our body cannot utilize the insulin after formation. According to the surveying data of WHO, 78 million people in India would be diabetic patients, which is ranked second among Asian countries next to China. In recent years Computer Science has provided different machine learning techniques for examining specific data sets. We observed that the data mining technique will be a very efficient pattern which specifies the dataset and gives valuable patterns to the applied classifiers. The purpose of this research is to identify or predict diabetes with the help of several machine learning classifiers. In this study we used WEKA software as mining tools for better diagnosis of diabetes. Here the PIMA INDIAN dataset is acquired from UCI repository. Our dataset is analyzed for the purpose of creating an effective model that may give the greatest precision and accuracy. In this research we also applied bootstrapping resample methodology to enhance the accuracy level. Furthermore, we used classifiers like naïve bayes, random forest, ANN, KNN, SVM and LR for accuracy. Also we analyze the comparative result along with these classifiers.
