Dimensionality reduction is an essential data preprocessing technique used for reducing the number of random variables under consideration. Traditional dimensionality reduction approaches fall under two categories one is Feature Extraction and Feature Selection. Many data warehouse contain massive amount of data, accumulated over long period of time. This is done in order to maintain the privacy concerns in which the data are aggregated to a certain levels. Another reason is to maintain a balance between the use of data that changes as data age and the size of data, thus avoiding over large data warehouse. In this paper an enhanced approach for business process has been carried out using dimensionality reduction by implementing wavelet transformation, which will help us in automated selection of most relevant independent attributes in a data warehouse.