Classifying of Diabetes Disease in Women Based on Support Vector Machine and Random Forest Algorithms

Abstract

Diabetes is a dangerous illness. It is indicated by blood sugar levels and/or glucose levels. Diabetes is a chronic illness that can lead to global health crisis, but there are things that can done to help control these crises. The primary energy source that get from the food on a daily basis in people with diabetes is blood sugar, or glucose. The insulin hormone created by the pancreas, assisting in the uptake of glucose from the blood into the cells so that it can be utilized as an energy source for daily tasks. Glucose remains in the blood when the body produces insufficient amounts of insulin, which can cause a number of health issues like heart attacks and strokes. There are numerous forms of diabetes, the most prevalent being type1 and type2. Type1 is typically diagnosed in children and young adults, whereas type 2 is typically found in middle age or older population. The objective of the project is to develop a system that can accurately classify patients as either diabetic or not by combining the results of various machine learning techniques with algorithms like Support Vector Machine and Random Forest. Machine learning is a scientific field where machine learning is derived from human experience. The model that best predicts diabetes is the one whose accuracy as a percentage is determined by calculating the accuracy of the model using each method. According to the experimental findings, RF and SVM have accuracy of 100% and 89% respectively and the precision of 100% and 91% respectively. Also, the recall (sensitivity) of RF and SVM are 100% and 95%, respectively.