Remote Healthcare Monitoring Architecture for Diabetes Patients Using Logistic Regression in Comparison Support Vector Machine for Improving Precision

The purpose of this project is to monitor remote healthcare and assist individuals with diabetes in making diabetic predictions. This study is divided into two groups. While the second group created a Support Vector Machine using 104 samples, the first group developed a Logistic Regression approach. With a Support Vector Machine (N = 52) and G power (value = 0.8) sampling approach, the sample size for Logistic Regression is 52. The Logistic Regression algorithm's accuracy is said to be 93%, whilst the Support Vector Machine Method's accuracy has been improved to 95%. The results of an independent sample T test indicated that the hypothesis is true and that the test is performed with an independent sample T test. The mean value was 93.2800, the mean accuracy detection was ±2SD, and the significance value was 0.000 (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$p&lt;0.05$</tex>). It was found that the end result of the Support Vector Machine Approach, 95%, was significantly more accurate than the Logistic Regression approach, 93%.