Fuzzy graph analysis for sociometry on latticed display

摘要

The fuzzy graph will make it possible to quantitatively analyze fuzzy information such as expression of human relations and mental processes. To clarify the main feature of a fuzzy graph, we would represent it as an approximate graph and extract its characteristics such as similarity and connectivity structures. Therefore we must proceed to analyze many kinds of information concerning the structure of a fuzzy graph, such as drawing and displaying comprehensively fuzzy graphs in the process. We have developed a computer-aided method for analyzing fuzzy graphs through human interaction. This method can quickly and comprehensively draw a graph arranged on a circular, a partition tree, cluster representative graph corresponding to a cluster and a specified shape of approximate n-valued fuzzy graph. However, it can not draw automatically a graph with nodes on the lattice intersection. So, we would rearrange the nodes on the lattice intersection for any fuzzy graph through user interface. This display is convenient for traditionally analyzing fuzzy graphs. We propose an analysis method using the lattice type of fuzzy graph display. Here, it plays an important role in this system. In this paper, we describe the analysis method and user interface of this system, and their application to sociometry analysis.