Data　clustering　is　a　popular　approach　for　automatically　finding　classes　or　groups　of　patterns.　In　recent　years,　data　clustering　is　still　a　popular　analysis　tool　for　data　statistics　to　identify　some　inherent　structures　that　presents　in　the　objects.　In　this　paper,　in　order　to　improve　the　convergence　and　global　searching　capacity　of　particle　swarm　optimization(PSO)　in　solving　data　clustering　,an　improved　particle　swarm　optimization　clustering　algorithm　(EPSOK)　based　on　reproductive　strategy　is　presented.　In　the　algorithm,　the　best　particles　in　the　search　process　reproduce,　at　the　same　time,　the　worst　particles　disappear.　Through　comparing　with　the　classical　K-Means　algorithm,　the　improved　algorithm　has　obvious　advantages　in　the　experiments.　©　2013　by　CESER　Publications.