Independent hybrid mining algorithms to increase the classification accuracy rates of decision tree (DT) and naïve Bayes (NB) classifiers for the classification of multi-class hitches. Both DT and NB classifiers are useful, proficient and commonly used for solving classification issues in data mining. In the second proposed hybrid NB classifier, we employ a DT instruction to select a reasonably more important subset of attributes for the production of native postulation of class conditional independence. Clustering is the task of consortium a set of instances in such a way that instances within a cluster have high similarities in comparison to one another, but are very divergent to instances in other clusters. The proposed algorithm also addresses some problems of data mining such as handling continuous attribute, dealing with missing characteristic values, and reducing noise in training data. Due to the large volumes of safekeeping audit data as well as the complex and dynamic assets of intrusion behaviors. Another difficulty of current IDS is to detect disturbances in real time high-speed networks, because the high-speed networks need IDS to deal with large volumes of network data in a very short time. The first proposed hybrid DT algorithm used a NB classifier to remove the noisy troublesome occasions from the training set before the DT induction, while the second proposed hybrid NB classifier used a DT induction to select a subset of features for the creation of naïve assumption of class restrictive independence.