Entropy approach through graph theory for studying the degree of order in one-dimensional distributions of objects
Armin Loeffler; Monique Rasigni; Uwe Raidt; Armin Loeffler; Département de Physique des Interactions Photons-Matière, case EC1, Faculté des Sciences et Techniques de St Jérôme, 13397 Marseille Cedex 20, France; Monique Rasigni; Département de Physique des Interactions Photons-Matière, case EC1, Faculté des Sciences et Techniques de St Jérôme, 13397 Marseille Cedex 20, France; Uwe Raidt; Département de Physique des Interactions Photons-Matière, case EC1, Faculté des Sciences et Techniques de St Jérôme, 13397 Marseille Cedex 20, France
Журнал:
Journal of Physics A: Mathematical and General
Дата:
1996-06-21
Аннотация:
Graph theory, through the minimal spanning tree (MST), and information theory, through the concept of entropy, are used to define a new parameter which quantitatively characterizes the degree of order (or disorder) in 1D sets of points. Theoretical calculations and results obtained from various computer-simulated distributions of points are compared. For well-chosen conditions which are specified, a good agreement is noted. Finally, a method which is particularly easy to implement is proposed to evaluate the parameter for any real or simulated 1D distribution of points.
144.8Кб