One of the most powerful technologies which are of high interest in the computer world is data mining. However, it is a comparatively new field of research whose major objective is to acquire knowledge from huge amounts of data. The motive of this study is to design a model which can forecast the likelihood of diabetes in patients with maximum accuracy. Therefore two main machine learning classification algorithms namely Decision Tree and Naive Bayes are used in this experiment to detect diabetes at an early stage. Experiments are performed on Pima Indians Diabetes Database (PIDD) which is sourced from the UCI machine learning repository. The performances of the two algorithms are evaluated on various measures like Precision, Accuracy, F-Measure, and Recall. Accuracy is measured over correctly and incorrectly classified instances. The filtering is done by comparing the group value to the class value followed by applying the Naïve Bayes classification algorithm for predicting diabetes. Significant attribute selection was done via the principal component analysis method. Results obtained show Naive Bayes outperforms with the highest accuracy of 76.30% compared to the decision tree which has 73.82%.