Understanding news story chains using information retrieval and network clustering techniques

Content analysis of news stories is a cornerstone of the communication studies field. However, much research is conducted at the level of individual news articles, despite the fact that news events are frequently presented as “stories” by news outlets: chains of connected articles offering follow up reporting as new facts emerge or covering the same event from different angles. These stories are theoretically highly important; they also create measurement issues for general quantitative studies of news output. Yet, thus far, the field has lacked an efficient method for detecting groups of articles which form stories in a way that enables their analysis. In this work, we present a novel, automated method for identifying news stories from within a corpus of articles, which makes use of techniques drawn from the fields of information retrieval and network analysis. We demonstrate the application of the method to a corpus of almost 40,000 news articles, and show that it can effectively identify valid story chains within the corpus. We use the results to make observations about the prevalence and dynamics of stories within the UK news media, showing that more than 50% of news production takes place within the form of story chains