Diabetes is a serious disease that spreads rapidly around the world and is first surprising. Its danger factors are therefore quite high. A different technique to identify diabetes at an early stage is to gradually decline in general health. This approach will reveal the health of the body’s organs before symptoms manifest. Using machine learning approaches, a framework for the diabetes prediction system is built in this research paper’s first stages. One popular method for streamlining the diabetes screening process is machine learning. A developed system using a machine learning algorithm and the PIMA health dataset. Using a health analyzer machine, health checks are performed. Diabetic and non-diabetic individuals are separated using the Support Vector machine (SVM) approach. The outcome displays the PIMA accuracy of the SVM algorithm.