An Efficient Estimation of diabetes mellitus using adapted SVM algorithm and cloud sanctuary

Global health issues like diabetes require early detection for care and prevention. A persistent metabolic disease with increase blood sugar stages might source stern health matters if not appropriately treated. Prevention of heart, kidney, and eye damage requires early and correct identification. This study introduces a diabetes prediction system using EPCA and MSVM to solve this need. EPCA reduces data dimensionality by keeping only the most important information, improving model prediction accuracy. For categorisation, MSVM, an advanced Support Vector Machine, is used. A powerful prediction model that accurately detects diabetes is created by combining EPCA with MSVM. The analysis framework uses Honeybee encryption to protect sensitive health data. This cloud-based encryption solution protects patient data from unauthorised access and breaches. Data integrity and secrecy are guaranteed using Honeybee encryption. EPCA-enhanced MSVM classifiers predict diabetes with improved accuracy, exceeding previous approaches. The framework's high accuracy allows healthcare providers to detect diabetes early and accurately for prompt intervention and control. Finally, this study offers a safe and accurate early diabetes prediction method. The approach secures patient data and improves prediction accuracy by integrating EPCA and MSVM with sophisticated encryption. This method helps doctors make diabetes-related decisions and improve patient outcomes.