期刊
INTERNATIONAL JOURNAL OF SPEECH TECHNOLOGY
卷 24, 期 3, 页码 601-614出版社
SPRINGER
DOI: 10.1007/s10772-021-09825-z
关键词
Diabetes mellitus; Random forest feature selection; Support vector regression; Logistic regression; Grid based random forest classifier; Natural language processing
The increasing prevalence of diabetes has led to the proposal of using machine learning for efficient diagnosis. By utilizing the ML Grid Search algorithm and the Pima Indian Diabetic Dataset, a diabetes prediction method with an accuracy of approximately 95.7% has been suggested, providing users with the ability to understand results in multiple languages.
Human body turns the food consumed into energy, but when insulin doesn't act in its way to convert the blood glucose into energy, then the glucose remains in the bloodstream and causes a life-threatening health issue called Diabetes Mellitus or Diabetes. According to the growing morbidity in recent years, in 2040, the world's diabetic patients will reach 642 million, which means that one of the ten adults in the future will be suffering from diabetes. With the rapid development of machine learning, it has been applied to many aspects of medical health. So, for efficiently and effectively diagnosing the Diabetes Mellitus, a method is proposed using the ML Grid Search algorithm. In this method, Pima Indian Diabetic Dataset is used. This system has two phases: The training phase includes preprocessing, feature selection and instance evaluation is done and the test phase includes preprocessing, instance evaluation and disease prediction is done. For feature selection, the random forest feature selection is used and for classification, support vector regression, logistic regression and grid based random forest classifier is used. The proposed method of predicting the diabetes, the accuracy is almost 95.7% which is higher when compared to previous methods. Additionally, the proposed system provides an ability to the users to understand the resulting scenario over any language with the help of language processing in results. The Natural Language Processing concept is adapted over the proposed approach to identify the features of the resulting text and perform the language processing and display the exact language-oriented data to the users without any hurdle.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据