This　paper　presents　Weight　Computing　in　Competitive　K-Means　Algorithm　which　is　derived　from　Improved　K-means　method　and　subspace　clustering.　By　adding　weights　to　the　objective　function,　the　contributions　from　each　feature　of　each　clustering　could　simultaneously　minimize　the　separations　within　clusters　and　maximize　the　separation　between　clusters.　The　experiments　described　in　this　paper　confirm　good　performance　of　the　proposed　algorithm.　©　2012　IEEE.