Spherical　k-means　clustering　is　a　generalization　of　k-means　problem　which　is　NP-hard　and　has　widely　applications　in　data　mining.　It　aims　to　partition　a　collection　of　given　data　with　unit　length　into　k　sets　so　as　to　minimize　the　within-cluster　sum　of　cosine　dissimilarity.　In　this　paper,　we　introduce　the　spherical　k-means　clustering　with　penalties　and　give　a　2　max{2,　M}(1　+　M)(ln　k　+　2)-approximate　algorithm,　where　M　is　the　ratio　of　the　maximal　and　the　minimal　penalty　values　of　the　given　data　set.