Abstract
Recognizing the importance of the Arabic publication for reaching a key target audience for ISIS, this expansive study of al-Naba’s text-based content will add to the scholarship of non-state actor communication using a mixed methodology of unsupervised machine learning and rhetorical analysis. First, it uses the ANTMN methodology to identify the frequency of the top Arabic thematic word clusters. Second, through time series and regression analysis, it adds to contemporary attention theories related to extremist group propaganda by uncovering what environmental factors correspond to changes in thematic emphasis. Third, it reveals the methodological steps necessary to improve evaluations of non-English large data discourse corpora.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
