This study investigates how international TV news channels present and frame current events in YouTube videos through combinations of multimodal resources. It further considers how multimodal analysis can be complemented by computational tools and approaches, and the trade-offs to be considered when dealing with both small and big data samples.
In the 21st century, international TV news is no longer the express purview of media giants like CNN or BBC. Today, news organisations around the world have established their own international news channels, providing viewers access to alternative views and perspectives on current global events. While broadcast television continues to be an enduring medium, audiences now have the option to access news content via the news channels' dedicated websites, or via YouTube, which has become an important distribution channel for international news networks. This trend engenders questions such as how international news organisations present events to their audiences, and how traditional news styles and formats have been adapted for this purpose. The present study has two aims: Using a multimodal social semiotic perspective, the first part of the paper investigates how pan-European niche-market international TV news channels such as DW News, Euronews, France 24, RT and TRT present current events through the complex multimodal interplay of audio, textual and visual data. Through the analysis of a small sample of select video clips released on the news organisations' official YouTube sites, the study highlights how meanings arise from the amalgam of language, text and images in dynamic news videos, and how these multimodal ensembles contribute to frame events, in this case, on the topic of 'Brexit', for international audiences. The second part of the paper investigates how conventional multimodal analysis can be complemented by automated video analytic tools, utilising computational linguistics, sentiment analysis and computer vision, such as offered by Google Cloud Video Intelligence or Microsoft Azure Video Indexer, for example. Comparing and contrasting traditional research methods with automated computational approaches, the paper highlights the insights that can be gained and trade-offs that need to be considered when dealing with both small samples and large corpora of news video data.
