Framing media coverage of the 2014 Sony Pictures entertainment hack: A topic modelling approach

Abstract

The Sony Pictures Entertainment hack of 2014, whiles not the biggest in the history of cyber invasion and terrorism, was however by far the most intensely covered in the news media. The political angle it assumed and the release of confidential data of some of the world's top celebrities, made it news worthy around the world. As the event brought focus on the issue of cyber terrorism, receiving traction in major news outlets around the world, the central theme as far as the hacking incident and its response in local and international media has not all been clear. Using the Latent Dirichlet Allocation (LDA) algorithm of automated topic modelling, this paper explores the essential themes in news coverage as reported in international and local media around the world. The approach is intended to extract interestingness measures in words usage in an attempt to inject semantic meaning into the narrative and vocabulary used in the media coverage of the incident. With large textual corpora, extracted on monthly basis, major news topics on the hacking incident in the US and the international media are compared. The results also help to create a monthly timeline story of events revealing inconsistencies as well as obvious similarities in the topics making news in both US and the rest of the world as far as the hacking event is concerned.