International Journal of Computer Science and Network
The present state-run of the fine art in the data linkage is to match the entities from the different data sources which do not contain the common identifier. In that here one-to-many data linkage is considered to obtain the decision process based on the clustering tree. In prior paper, there is no one-to-many data linkage tasks instead the issue addressed are to link among the same type of the entities. In this paper, two new splitting criterion are introduced to enhance the performance of the linkage process for the best split at each node during the decision tree construction process and securing the linked data from the unauthorized usage.