Anzeige
Erschienen in:
2013 | OriginalPaper | Buchkapitel
Properties of a General Measure of Configuration Agreement
verfasst von : Stephen L. France
Erschienen in: Algorithms from and for Nature and Life
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Variants of the Rand index of clustering agreement have been used to measure agreement between spatial configurations of points (Akkucuk 2004; Akkucuk and Carroll 2006; Chen 2006; Chen and Buja 2009). For these measures, the k-nearest neighbors of each point are compared across configurations. The agreement measure can be generalized across multiple values of k (France and Carroll 2007). The generalized agreement metric is denoted as ψ. In this paper, we further generalize ψ to the case of more than two configurations. We develop a partial agreement measure as a neighborhood agreement equivalent of the partial correlation coefficient. We demonstrate the use of ψ and partial ψ using an illustrative example. (MATLAB implementations of routines for calculating ψ and partial ψ are available at https://sites.google.com/site/psychminegroup/.)