Despite　many　successful　stories　of　data　mining　in　a　wide　range　of　applications,　this　technique　has　raised　some　issues　related　to　privacy　and　security　of　individuals.　Due　to　these　issues,　data　owners　are　often　unwilling　to　share　their　sensitive　information　with　data　miners.　In　this　paper,　we　present　a　novel　method　for　privacy　preserving　clustering　over　centralized　data.　The　proposed　method　is　built　upon　the　application　of　Double-Reflecting　Data　Perturbation　Method　(DRDP)　and　Rotation　Based　Translation　(RBT)　in　order　to　provide　secrecy　of　confidential　numerical　attributes　without　losing　accuracy　in　results.　The　experiments　demonstrate　that　the　proposed　method　is　effective　and　provides　a　feasible　approach　to　balancing　privacy　and　accuracy.　©2009　IEEE.