A SOCIAL NETWORK ANALYSIS APPROACH TO ANALYZE ROAD NETWORKS

Abstract

Depending on relations and transient properties, real world events or physical settings can be represented with different types of network structures. To realize this purpose, a social network topology becomes a complex and a scale free network. In this paper, we propose to use this topology to represent road networks. Generally speaking, road networks and their spatial relations reside on a plane, which generates a specific network structure termed as a planar network. This observations result in the degree distributions in road networks which contain a considerably different characteristic compared to other types of networks. In order to analyze these special networks, we introduce a novel approach using the centrality and entropy of various distributions estimated from the network topology. In context of sociology, the centrality is used for find an important actor in a social setting. The entropy of these centrality distributions provides a way to examine the characteristics of selected road networks at city, county or state levels. We conjecture that the proposed tools will help transportation experts to gather in depth information on existing road networks. Our experiments prove a concept design, which can be used to quantitatively compare residential area and downtown area road networks.