White house under attack: Introducing distributional semantic models for the analysis of US crisis communication strategies

In the present contribution, the use of distributional semantic models (DSMs) is proposed as a novel approach for the analysis of crisis communication strategies. Temporal Random Indexing (TRI), a specific DSM framework, is employed as computational tool to analyse word meaning change over time. Our resource is represented by the CompWHoB (Computational White House press Briefings) Corpus, a political diachronic corpus collecting the transcripts of the White House Press Briefings from 1993 to 2014. Primary objective of this paper is to demonstrate that TRI can be used in conjunction with Critical Discourse Analysis (CDA) theories as an easily adaptable tool applicable to the analysis of the so-called crisis communication management moments where US administration has to cope with risky and serious scenarios.