Understanding the network fundamentals of news sources associated with a specific topic

Most real-world networks such as the World Wide Web and social networks are well understood by means of social network analysis techniques, and most of these networks are scale-free rather than random. In this article, we propose a novel network of news sources, who are influential persons who appear in news articles. We first collect a group of news articles associated with a particular subject (e.g., abortion). Then, we identify and extract all news sources from the group of the news articles. In the news source network, news sources are represented as nodes, and two nodes are connected to each other if the news sources that correspond to the nodes appear in the same news article. Given a news source network, the goal of our research is to determine the underlying principles of the network. Our experimental results indicate that news source networks are a scale-free network with a heavy-tailed distribution because of the burst property of news articles. In addition, degree and betweenness measures are useful to identify the influential persons of a particular topic. Unlike most existing social networks, news source networks tend to show an interesting pattern, which is the polarity of high and low scores in both closeness and clustering coefficient measures. Moreover, news source networks do not always seem to have a gigantic component. Finally, we present how our proof-of-concept prototype system is implemented with a new graph layout algorithm modified from the Fruchterman and Reingold algorithm.