Design and Implementation of a Diabetes Prediction Model With Meal and Lifestyle Recommendation

Diabetes mellitus is a major global health issue, with rising prevalence due to poor lifestyle habits and lack of early diagnosis. This project focuses on developing an intelligent system for early diabetes prediction and general health guidance. The aim is to enhance proactive diabetes management through accurate diagnosis and lifestyle recommendations. The system was developed using the Support Vector Machine (SVM) algorithm, trained on a well-structured clinical dataset containing parameters such as glucose level, BMI, age, gender, heart disease, and blood pressure. Preprocessing involved data normalization to improve model performance. The model was evaluated using accuracy, precision, recall, and F1-score, demonstrating high predictive capability in classifying diabetic and non-diabetic individuals. A user-friendly web application was built using Python and Flask, allowing users to input health data and receive instant feedback along with general dietary and lifestyle recommendations. The findings show that combining machine learning with practical health advice can effectively support early diagnosis and empower users in managing diabetes.