Exploring TV Seriality and Television Studies through Data-Driven Approaches

Abstract

The chapter discusses the use of data-driven approaches in television studies, which has become possible due to the increasing availability of digital data. Computational techniques can be used to analyze cultural artifacts, gain insights into audience reactions to specific shows or episodes, and investigate patterns and trends in television programming over time. The chapter also highlights the challenges of analyzing television series, which are complex open systems that interact with external factors such as the production process, audience feedback, and cultural and social context. Content analysis, which involves qualitative and quantitative methods based on the coding process and data collection, can be used to analyze various elements of a TV series.

Generative AI is also discussed, which refers to the use of deep multi-modal algorithms to generate new content such as images, speech, and text. Generative methods like Generative Adversarial Networks (GANs) and Stable Diffusion can create new content that is almost indistinguishable from real data. While generating videos is more challenging, Recurrent Neural Networks (RNNs) like LSTMs can capture the temporal dynamics of the scenes to create interesting and promising applications for complex, but short-duration videos.

Keywords

Television series; television studies; data-driven audiovisual analysis; generative AI; content analysis.

Download Chapter