Analysis of Diabetes Disease Prediction Using Machine Learning Algorithms

Diabetes is a frequent health problem caused by pancreatic dysfunction. It raises blood sugar levels and has serious consequences. A person suffering from these disease experiences severe symptoms affecting the kidneys, heart, eyes, etc. A substantial amount of investigation shows that machine learning (ML) algorithms can enhance sickness detection and minimize medical mistake rates, potentially saving life of an individual. It's often challenging to build a reliable and impactful predictive model because of absent values and outliers in the data frame. Using a dataset with outliers and missing values due to class imbalance and the intent of this work is to develop Diabetes Prediction (DP) & classification Model whose performance is accurate and robust. At the Pima Indian Diabetes (PID) data frame derived from the kaggel diabetes data frame, a pre processing technique will be applied. This study includes five Algorithms to calculate the model's accuracy such as Support Vector Machine (SVM), Navie bayes (NB), Random Forest (RF), K- Nearest Neighbor (KNN), Decision Tree (DT). The most accurate model is then used to forecast diabetes.