GUI-Based Diabetes Prediction Using Pipeline

Diabetes is a global health problem that hinders health progress. This condition can be detected early for improved health and longer life. Supervised machine learning algorithms trained on adequate datasets can diagnose diabetes early. We discover dataset features and evaluate a GUI-based diabetes prediction system utilizing a pipeline to give a thorough guide to predicting diabetic individuals. GUI-based diabetes prediction systems that employ pipelines require a graphical user interface to enter health data including age, weight, and blood sugar levels. After passing this data through a pipeline of machine learning algorithms, the user's diabetes risk would be predicted. This research trained datasets using machine learning algorithms including KNN, DT, RF, GB, SVM, and LightGBM Classifier. Reprocessing, training, visualization, and analysis are common pipeline steps. The algorithm with the best results is used to forecast fresh user input data via the GUI. The GUI would show the projected diabetes status. The study's findings show that diabetes can be predicted with accuracy and efficiency by applying machine learning-based classification and properly pre-processing clinical data.