Abstract
Many people are threatened with diabetes, making it one of the most serious diseases nowadays. Diabetes can be brought on by many factors, including advancing age, being overweight, not getting enough exercise, having a family history of the disease, leading an inactive lifestyle, eating improperly, having high blood pressure, etc. Diabetics are at increased risk for a wide range of health issues, including cardiovascular disease, kidney failure, stroke, vision and nerve problems, and more. Diabetic patients typically undergo a series of diagnostic examinations at the hospital before receiving therapy, as this is the standard procedure. The healthcare industry is a key user of big data analytics. The healthcare sector makes use of massive data sets. The study of massive datasets using big data analytics allows one to uncover previously unknown information and patterns, allowing one to draw conclusions and make predictions about the data. To improve diabetes classification, this research proposes a hybrid prediction model that combines the strengths of two Machine Learning (ML) models: the Support Vector Machine (SVM) and the Random Forest (RF). To evaluate the efficacy of the proposed hybrid model, the results of the model were compared with simple ML models, including SVM, RF, and Decision Tree (DT). The hybrid model with good accuracy is finalized and deployed in the web application. The website's layout makes it simple for visitors to navigate and use the interface to determine their health conditions.
Keyword
Diabetes, Data Processing, Feature Extraction, Machine Learning, Accuracy, Website
PDF Download (click here)
|