An Efficient Approach of Discovery of Frequent Data Set from Big Operational Database

Abstract

Currently in real world scenario data uncertainty is the most major issue in the real time applications where these data are generated from various devices daily from various users. So, the important part is to find the important data from them. In this paper, we propose to measure pattern frequentness based on the various possible world semantics. We are looking to establish two uncertain sequence data models abstracted from many real-life applications involving uncertain sequence data, and based on that formulate the problem of mining probabilistically frequent sequential patterns (or p-FSPs) from data that conform to our models. By using the projection strategy of famous prefixspan algorithm, we are looking to develop an algorithm called U-PrefixSpan for probabilistically frequent sequential pattern mining. UPrefixSpan avoids the problem of “possible world explosion” and when combined with pruning techniques and one validating technique achieves good performance. Theoretically study and analysis shows that our work proposed do the better with compare to existing system.
DOI: 10.17762/ijritcc2321-8169.150786