Data mining is the study of data and the utility of software techniques for finding patterns and regularities in sets of data. The main reason behind the paper is to examine the exhibition of an existing client in order to the churning prediction to actualize bunching with the characterization expectation model including weighted K-means clustering with the Neural system. To assess and break down the exhibition based on exactness, accuracy, review, root means square mistake and to analyze the presentation of the proposed strategy with the current cross breed strategies. For this reason, numerous algorithms have been proposed to anticipate churning clients. In the proposed work, Decision Tree and Neural systems are utilized to break down the examples from the dataset and to hybrid the model for customer behavioral mining. As per the current system, both the K-means and Neural system have been utilized either independently or with the other mix too. It was seen that every procedure will give better outcomes. However, the issue that has been engaged in the investigation is to upgrade the k-implies calculation. A weighted K-means is used to improve the exactness of the classifier and utilized to distinguish agitate clients and compelling promoting systems. Now, these days customers for seeking satisfaction shifts from one homogeneous product to another termed as customer churning which is proposed through various techniques through this paper.