Preserving privacy of transactional data which is sensitive in the real world. Database consisting of large datasets with high dimensionality involves huge number of individual records. It is necessary to generalize the data such that it ensures privacy. In the existing technique, paradigm such as K-anonymity is used to publish data such that it is not transparent to draw any inference out of it. Even though, it is efficient for fixed schema and low dimensional data it is not effective for certain applications which include high dimensional data.