Finding influential sources and breaking news in news media using graph analysis techniques

The popularity of social media has increased the interest for knowledge extraction from social networks and social media sites. The discovery of influential content or users and hidden social connections can be profitable for social media users and companies through personalisation and promotion respectively. Despite the abundance of works on social media and networks, there are no similar works in traditional (i.e., press, radio, TV) or online media (i.e., news sites). This work proposes a solution that solves the lack of influence or connection information by analysing news media content. Consequently, it detects the underlying influence among news media companies and provides knowledge about breaking news. Among the contributions of this work are: a new methodology for identifying and quantifying the implicit influence between news media, based on content similarity and a new method for the early detection of breaking news, with high interest to the mass media.