In this research, I would like to introduce two independent hybrid machine learning algorithms to improve the accuracy rates of AdaBoost and naïve Bayes (NB) classifiers for the classification of noisy and high dimensional data’s. Both AdaBoost and NB classifiers are useful, efficient and commonly used for solving classification problems in data mining. Since the training dataset may contain lots of missing values, noises and contradictory instances, if someone try to run AdaBoost classifier on that training data, at the time of generation of decision trees, the decision trees might suffer from overfitting and its accuracy might decrease. Now, in the first proposed hybrid ADA+NB classifier, we employ a AdaBoost induction to select comparatively more important subset of attributes for the production of naïve assumption of class conditional independence. In my second proposed hybrid NB+ADA classifier, I would like to employ a naïve Bayes (NB) classifier to remove the noisy troublesome instances from the training set before the DT induction at AdaBoost classifier. We tested the performances of the two proposed hybrid algorithms against the existing AdaBoost and NB classifiers respectively using the classification accuracy, precision, sensitivity, specificity analysis, and 70% by 30% split validation on some real benchmark datasets from UCI (University of California, Irvine) machine learning repository who are very noisy and comparatively high dimensional.