Incorporation of Business Intelligence on Uncertain Events

Uncertainty is a state of lack of certainty, where having incomplete information can make system impossible to describe the desired outcome. Uncertainty of data is inevitable in every database in which data changes takes place frequently. As models of the real world, databases are often permeated with forms of uncertainty, including incompleteness, ambiguity, vagueness, inconsistency and imprecision. There is need to remove uncertainty in databases. Automatic elimination of uncertainty would make a real time application more efficient and reduce the human work. In this paper, we apply the Business Intelligence technique called k-means clustering and Bayesian network for reduction of uncertainty in database. The framework of Business Intelligence on Uncertain Events (BI_UnEvent) processes the uncertain events efficiently. Experimental results on the alumni database of Pondicherry Engineering College (PEC) data set show the novelty of our approach in removing uncertainty. It is observed that the proposed technique performs better in terms of processing time and precision when applied to College alumni database data sets compared to a Bayesian Network based approach.