Neural Network Classification to Determine the Likelihood of Diabetes Using Python Programming Language

Abstract

Diabetes is a global health problem that affects millions of people worldwide. Predicting a person's risk of developing diabetes can be an important first step in disease prevention and management. In this study, we propose the development of a predictive model for diabetes using Neural Network (NN) technique with implementation using Python. The data used in this study consists of clinical information that includes factors such as pregnancy, glucose, blood pressure, skin thickness, insulin, BMI, diabetes pedigree function, and age. The model development process involves data pre-processing, selection of relevant features, model training, and performance evaluation using appropriate metrics. The experimental results show that the developed NN model has a good ability in predicting diabetes risk. The main contribution of this research is the use of NN techniques and Python coding in the development of predictive models for diabetes, which can provide useful guidance for medical practitioners in supporting disease prevention and management efforts. Future studies can extend this research by considering additional factors and improving the accuracy of the model by using more complex approaches. Keywords: Diabetes, Prediction, Neural Network, Python coding, Predictive model, Model development, Data pre-processing, Performance evaluation, Disease prevention, Disease management