Web graph analysis of the air traffic safety community

Abstract

The structure and properties of the web graph originating from the air traffic safety community could reveal many interesting relationships which exist between key organizations in the field. Therefore, an attempt is made to collect a representative sample of relevant web pages from different sources, by utilizing an intelligent software agent capable of recognizing the concept of a typical air traffic safety page. Starting from a seed set of thirty relevant pages, the agent collected nearly 200,000 on-topic pages and recorded the links between them. The analysis of the compiled graph detected the most relevant pages and sites, common linking patterns and popular keywords used for titles and links texts.