Introduction: At this junction, machine learning demand is enhancing in almost every critical area to catch interesting and decision-making patterns. This inductive research objective is investigating sophisticated different techniques of machine learning to effectively analyze health data. Naturally, the present health-related dataset is most sensitive, crucial, and needs accurate analysis, hence result generated by different learning algorithms have paramount importance. This sensitivity enhances and promoted data analytics, interest, and role through machine learning in the health sector. Objectives: This research aims to analyze and predict diabetes by applying elegant learning algorithms on the diabetes dataset. The article also shows a comparative study analysis of algorithms. Methods: This research uses the median method to preprocess the dataset. After preprocessing, ten different machine learning algorithms are applied to the diabetes dataset in this paper. Results: This document uses a diabetes dataset that has eight different symptoms or features to predict disease. To get a better classification technique, various ML mechanisms results are compared and analyzed. This study outcome can be further utilized in incoming research based on diabetic health problems. Conclusion: A linear support vector machine shows better detection results compared to others.