An incremental nonsmooth optimization algorithm for clustering using L_1 and L_\infty norms

Abstract

<p style='text-indent:20px;'>An algorithm is developed for solving clustering problems with the similarity measure defined using the <inline-formula><tex-math id="M3">\begin{document}$ L_1 $\end{document}</tex-math></inline-formula> and <inline-formula><tex-math id="M4">\begin{document}$ L_\infty $\end{document}</tex-math></inline-formula> norms. It is based on an incremental approach and applies nonsmooth optimization methods to find cluster centers. Computational results on 12 data sets are reported and the proposed algorithm is compared with the <inline-formula><tex-math id="M5">\begin{document}$ X $\end{document}</tex-math></inline-formula>-means algorithm.