%0 Generic %A Fu, Wei %A Perry, Patrick O. %D 2019 %T Estimating the number of clusters using cross-validation %U https://tandf.figshare.com/articles/dataset/Estimating_the_number_of_clusters_using_cross-validation/9034343 %R 10.6084/m9.figshare.9034343.v1 %2 https://tandf.figshare.com/ndownloader/files/16526753 %2 https://tandf.figshare.com/ndownloader/files/16526756 %2 https://tandf.figshare.com/ndownloader/files/16526759 %2 https://tandf.figshare.com/ndownloader/files/16526762 %2 https://tandf.figshare.com/ndownloader/files/16526765 %2 https://tandf.figshare.com/ndownloader/files/16526768 %2 https://tandf.figshare.com/ndownloader/files/16526771 %2 https://tandf.figshare.com/ndownloader/files/16526774 %2 https://tandf.figshare.com/ndownloader/files/16526777 %2 https://tandf.figshare.com/ndownloader/files/16526780 %2 https://tandf.figshare.com/ndownloader/files/16526783 %2 https://tandf.figshare.com/ndownloader/files/16526786 %2 https://tandf.figshare.com/ndownloader/files/16526789 %2 https://tandf.figshare.com/ndownloader/files/16526792 %2 https://tandf.figshare.com/ndownloader/files/16526795 %2 https://tandf.figshare.com/ndownloader/files/16526798 %2 https://tandf.figshare.com/ndownloader/files/16526801 %2 https://tandf.figshare.com/ndownloader/files/16526804 %2 https://tandf.figshare.com/ndownloader/files/16526807 %2 https://tandf.figshare.com/ndownloader/files/16526810 %2 https://tandf.figshare.com/ndownloader/files/16526813 %2 https://tandf.figshare.com/ndownloader/files/16526816 %2 https://tandf.figshare.com/ndownloader/files/16526819 %2 https://tandf.figshare.com/ndownloader/files/16526822 %2 https://tandf.figshare.com/ndownloader/files/16526825 %2 https://tandf.figshare.com/ndownloader/files/16526828 %2 https://tandf.figshare.com/ndownloader/files/16526831 %2 https://tandf.figshare.com/ndownloader/files/16526834 %2 https://tandf.figshare.com/ndownloader/files/16526837 %2 https://tandf.figshare.com/ndownloader/files/16526840 %2 https://tandf.figshare.com/ndownloader/files/16526843 %2 https://tandf.figshare.com/ndownloader/files/16526846 %2 https://tandf.figshare.com/ndownloader/files/16526849 %2 https://tandf.figshare.com/ndownloader/files/16526852 %2 https://tandf.figshare.com/ndownloader/files/16526855 %2 https://tandf.figshare.com/ndownloader/files/16526858 %2 https://tandf.figshare.com/ndownloader/files/16526861 %2 https://tandf.figshare.com/ndownloader/files/16526864 %2 https://tandf.figshare.com/ndownloader/files/16526867 %2 https://tandf.figshare.com/ndownloader/files/16526870 %2 https://tandf.figshare.com/ndownloader/files/16526873 %2 https://tandf.figshare.com/ndownloader/files/16526876 %2 https://tandf.figshare.com/ndownloader/files/16526879 %2 https://tandf.figshare.com/ndownloader/files/16526882 %2 https://tandf.figshare.com/ndownloader/files/16526885 %2 https://tandf.figshare.com/ndownloader/files/16526888 %2 https://tandf.figshare.com/ndownloader/files/16526891 %2 https://tandf.figshare.com/ndownloader/files/16526894 %2 https://tandf.figshare.com/ndownloader/files/16526897 %2 https://tandf.figshare.com/ndownloader/files/16526900 %2 https://tandf.figshare.com/ndownloader/files/16526903 %2 https://tandf.figshare.com/ndownloader/files/16526906 %2 https://tandf.figshare.com/ndownloader/files/16526909 %2 https://tandf.figshare.com/ndownloader/files/16526912 %2 https://tandf.figshare.com/ndownloader/files/16526915 %2 https://tandf.figshare.com/ndownloader/files/16526918 %2 https://tandf.figshare.com/ndownloader/files/16526921 %2 https://tandf.figshare.com/ndownloader/files/16526924 %2 https://tandf.figshare.com/ndownloader/files/16526927 %2 https://tandf.figshare.com/ndownloader/files/16526930 %2 https://tandf.figshare.com/ndownloader/files/16526933 %2 https://tandf.figshare.com/ndownloader/files/16526936 %2 https://tandf.figshare.com/ndownloader/files/16526939 %2 https://tandf.figshare.com/ndownloader/files/16526942 %2 https://tandf.figshare.com/ndownloader/files/16526945 %2 https://tandf.figshare.com/ndownloader/files/16526948 %2 https://tandf.figshare.com/ndownloader/files/16526951 %2 https://tandf.figshare.com/ndownloader/files/16526954 %2 https://tandf.figshare.com/ndownloader/files/16526957 %2 https://tandf.figshare.com/ndownloader/files/16526960 %2 https://tandf.figshare.com/ndownloader/files/16526963 %2 https://tandf.figshare.com/ndownloader/files/16526966 %2 https://tandf.figshare.com/ndownloader/files/16526969 %2 https://tandf.figshare.com/ndownloader/files/16526972 %2 https://tandf.figshare.com/ndownloader/files/16526975 %2 https://tandf.figshare.com/ndownloader/files/16526978 %2 https://tandf.figshare.com/ndownloader/files/16526981 %2 https://tandf.figshare.com/ndownloader/files/16526984 %2 https://tandf.figshare.com/ndownloader/files/16526987 %2 https://tandf.figshare.com/ndownloader/files/16526990 %2 https://tandf.figshare.com/ndownloader/files/16526993 %2 https://tandf.figshare.com/ndownloader/files/16526996 %2 https://tandf.figshare.com/ndownloader/files/16526999 %2 https://tandf.figshare.com/ndownloader/files/16527002 %2 https://tandf.figshare.com/ndownloader/files/16527005 %2 https://tandf.figshare.com/ndownloader/files/16527008 %2 https://tandf.figshare.com/ndownloader/files/16527011 %2 https://tandf.figshare.com/ndownloader/files/16527014 %2 https://tandf.figshare.com/ndownloader/files/16527017 %2 https://tandf.figshare.com/ndownloader/files/16527020 %2 https://tandf.figshare.com/ndownloader/files/16527023 %2 https://tandf.figshare.com/ndownloader/files/16527026 %2 https://tandf.figshare.com/ndownloader/files/16527029 %2 https://tandf.figshare.com/ndownloader/files/16527032 %2 https://tandf.figshare.com/ndownloader/files/16527035 %2 https://tandf.figshare.com/ndownloader/files/16527038 %2 https://tandf.figshare.com/ndownloader/files/16527041 %2 https://tandf.figshare.com/ndownloader/files/16527044 %2 https://tandf.figshare.com/ndownloader/files/16527047 %2 https://tandf.figshare.com/ndownloader/files/16527050 %2 https://tandf.figshare.com/ndownloader/files/16527053 %2 https://tandf.figshare.com/ndownloader/files/16527056 %2 https://tandf.figshare.com/ndownloader/files/16527059 %2 https://tandf.figshare.com/ndownloader/files/16527062 %2 https://tandf.figshare.com/ndownloader/files/16527065 %2 https://tandf.figshare.com/ndownloader/files/16527068 %2 https://tandf.figshare.com/ndownloader/files/16527071 %2 https://tandf.figshare.com/ndownloader/files/16527074 %2 https://tandf.figshare.com/ndownloader/files/16527077 %2 https://tandf.figshare.com/ndownloader/files/16527080 %2 https://tandf.figshare.com/ndownloader/files/16527083 %2 https://tandf.figshare.com/ndownloader/files/16527086 %2 https://tandf.figshare.com/ndownloader/files/16527089 %2 https://tandf.figshare.com/ndownloader/files/16527092 %2 https://tandf.figshare.com/ndownloader/files/16527095 %2 https://tandf.figshare.com/ndownloader/files/16527098 %2 https://tandf.figshare.com/ndownloader/files/16527101 %2 https://tandf.figshare.com/ndownloader/files/16527104 %2 https://tandf.figshare.com/ndownloader/files/16527107 %2 https://tandf.figshare.com/ndownloader/files/16527110 %2 https://tandf.figshare.com/ndownloader/files/16527113 %2 https://tandf.figshare.com/ndownloader/files/16527116 %2 https://tandf.figshare.com/ndownloader/files/16527119 %2 https://tandf.figshare.com/ndownloader/files/16527122 %2 https://tandf.figshare.com/ndownloader/files/16527125 %2 https://tandf.figshare.com/ndownloader/files/16527128 %2 https://tandf.figshare.com/ndownloader/files/16527131 %2 https://tandf.figshare.com/ndownloader/files/16527134 %2 https://tandf.figshare.com/ndownloader/files/16527137 %2 https://tandf.figshare.com/ndownloader/files/16527140 %2 https://tandf.figshare.com/ndownloader/files/16527143 %2 https://tandf.figshare.com/ndownloader/files/16527146 %2 https://tandf.figshare.com/ndownloader/files/16527149 %2 https://tandf.figshare.com/ndownloader/files/16527152 %2 https://tandf.figshare.com/ndownloader/files/16527155 %2 https://tandf.figshare.com/ndownloader/files/16527158 %2 https://tandf.figshare.com/ndownloader/files/16527161 %2 https://tandf.figshare.com/ndownloader/files/16527164 %2 https://tandf.figshare.com/ndownloader/files/16527167 %2 https://tandf.figshare.com/ndownloader/files/16527170 %2 https://tandf.figshare.com/ndownloader/files/16527173 %2 https://tandf.figshare.com/ndownloader/files/16527176 %2 https://tandf.figshare.com/ndownloader/files/16527179 %2 https://tandf.figshare.com/ndownloader/files/16527182 %2 https://tandf.figshare.com/ndownloader/files/16527185 %2 https://tandf.figshare.com/ndownloader/files/16527188 %2 https://tandf.figshare.com/ndownloader/files/16527191 %2 https://tandf.figshare.com/ndownloader/files/16527194 %2 https://tandf.figshare.com/ndownloader/files/16527197 %2 https://tandf.figshare.com/ndownloader/files/16527200 %2 https://tandf.figshare.com/ndownloader/files/16527203 %2 https://tandf.figshare.com/ndownloader/files/16527206 %2 https://tandf.figshare.com/ndownloader/files/16527209 %2 https://tandf.figshare.com/ndownloader/files/16527212 %2 https://tandf.figshare.com/ndownloader/files/16527215 %2 https://tandf.figshare.com/ndownloader/files/16527218 %2 https://tandf.figshare.com/ndownloader/files/16527221 %2 https://tandf.figshare.com/ndownloader/files/16527224 %K clustering %K cross-validation %K k-means %K model selection %K unsupervised learning %X

Many clustering methods, including k-means, require the user to specify the number of clusters as an input parameter. A variety of methods have been devised to choose the number of clusters automatically, but they often rely on strong modeling assumptions. This paper proposes a data-driven approach to estimate the number of clusters based on a novel form of cross-validation. The proposed method differs from ordinary cross-validation, because clustering is fundamentally an unsupervised learning problem. Simulation and real data analysis results show that the proposed method outperforms existing methods, especially in high-dimensional settings with heterogeneous or heavy-tailed noise. In a yeast cell cycle dataset, the proposed method finds a parsimonious clustering with interpretable gene groupings.

%I Taylor & Francis