Diabetes is a metabolic and chronic disease which is described by the prominent levels of blood glucose levels. Type II diabetes is the most common form of diabetes. This chapter proposes an improved combination of Particle Swarm Optimization algorithm with decision trees for the assessment of risk factors that correspond to type II diabetes. Data optimization methods are deployed in various fields such as engineering, medical sciences, management, physical sciences, and social learning. The process behind optimization is to choose the best solution from the set of feasible solutions thereby providing scientific decision-making. The modules corresponding to the proposed workflow are data collection, data pre-processing, feature selection, classification, and model evaluation. Data collection involves the retrieval of dataset related to type II diabetes such as patient’s data, blood sample data, and its comorbidities. The within-class scatter matrix is proportional to the sample covariance for the pooled dimensional data.