A Survey on Diabetes Prediction Using Machine Learning Techniques

Diabetes is one of the most common chronic diseases worldwide and poses a serious threat to 
public health. Early prediction of diabetes can help in timely diagnosis and prevention of severe 
complications. Traditional diagnostic methods often rely on clinical tests and expert analysis, 
which can be time-consuming and expensive. In recent years, machine learning and artificial 
intelligence techniques have been widely used to improve the accuracy and efficiency of 
diabetes prediction systems. 
This survey paper reviews recent research (2019–2025) on diabetes prediction using machine 
learning and deep learning approaches. Various models such as Decision Trees, Support Vector 
Machines, Random Forest, Neural Networks, and ensemble techniques are analyzed and 
compared based on their performance, accuracy, and applicability. The study also discusses 
challenges such as data quality, model complexity, and real-world implementation issues. It is 
observed that machine learning-based systems significantly improve prediction accuracy, but 
further research is required to develop cost-effective and scalable healthcare solutions.