Early Diabetes Prediction using Random Forest

Non-communicable diseases such as diabetes and hypertension account for nearly half of all early deaths every year. Diabetes is a disease that necessitates continuous monitoring of blood sugar levels, particularly glucose, in order to effectively mitigate the emerging health complications. Self-management of the disease, particularly maintaining glucose levels in the blood within a specific range, is critical for treatment. This involves actively monitoring blood glucose and insulin levels, as well as managing diet and physical activity. Diabetes advancements and self-administration applications have made it easier for patients to gain access to more relevant information. The ability of Machine Learning (ML) techniques to solve complex tasks in dynamic environments and with dynamic data has contributed to its success in the study of diseases such as diabetes. This research study has attempted to design and develop a personal healthcare system, which can use the data generated from daily activities and other significant parameters to monitor the blood glucose level and provide proper report representation online. The Machine Learning (ML) technique has been used to detect non-communicable diseases at an early stage. The proposed research study has surveyed previous relevant literatures in order to understand the approaches used previously for the detection or presence of disease in the human body. There are various methods to determine blood glucose levels non-invasively, but a universally acceptable method with good accuracy and precision is not yet available. This paper emphasized about using Random Forest, a supervised machine learning algorithm to determine the probability of occurrence of diabetes mellitus to take precautions accordingly.