Difference in performance of the Naive bayes and AODE algorithms




I read that like naive Bayes, AODE does not perform model selection and does not use tuneable parameters. As a result, it has low variance. It predicts class probabilities rather than simply predicting a single class, and can also handle the missing data.
Which of the two algorithms performs good? Is there a difference in situations where we should apply them or that one would outperform the other every time?



@pravin According to my knowledge naive bayes alogorithm is applicable only when the predictors are independent such as Suppose if a fruit has shape is round,color is red and diameter is 3 cm then we will call that fruit as Apple so we can apply Naive Bayes Algorithm only when all the attribute of apples are independent in the sentence that means presence of round shape is not affected by color red or having diameter 3 cm. In that only Naive Bayes Classifier can be used but i dont know AODE Algorithm Could you please explain more about AODE?