Extended PSD Estimation by Using Maximum Clique Enumeration

抄録

This paper proposes a novel PSD estimation to estimate admissible value intervals of missing elements from the limited known elements in a Positive Semi-Definite (PSD) similarity matrix where the known part of the matrix represents a joined complete split graph. The main contribution of this study is the extension of the applicability and the accuracy of the PSD estimation by introducing a maximal clique enumeration.