Data-driven storytelling videos : cinematic guidelines and design space

Abstract

Storytelling is a skill that humans have developed throughout their evolutionary journey to meet the needs of information transmission and communication. Crafting a compelling data story demands more attention and study in the big data era. Videos combine the art of storytelling with cinematic audiovisual elements in a time-based narrative media that effectively deliver information to the general public. Many data-driven videos are utilized for promotional, social, and political purposes posing essential questions and challenges in society, academia, and industry. In this thesis, I meticulously investigate this phenomenon and the ensuing research problems via a systematic combination of research methods, including corpus analysis, expert interviews, and user studies grounded in an interdisciplinary methodology that integrates cinematic arts with data visualization. Specifically, this thesis centers on the following: (1) Establishing design guidelines for data videos. Given the relatively short history of data video creation, there is a dearth of extensive creative experience. The opening and ending of a data video significantly influence its attractiveness and impressiveness. To augment the effectiveness of data video creation and facilitate accurate and efficient information transmission to the general audience, I merge established cinematic techniques to study and develop design guidelines applicable to data videos. (2) Exploring the design space for short technical storytelling videos. Many short technical storytelling videos, such as preview videos at VIS conferences, provide valuable exemplars of a design space. I identify common design patterns in the context of visualization conferences and correlate them with the storytelling design considerations of movie trailers and television advertisements. This thesis contributes to the interdisciplinary study bridging cinematic arts and data visualization by integrating cinematic storytelling and data-driven storytelling videos. The aim is to assist data designers in conveying data insights to the general public in an efficient and intuitive manner. I aspire that this interdisciplinary study will continue to inspire more engaging and effective storytelling techniques for data-driven storytelling videos and other storytelling mediums.

Date of Award	2023
Original language	English
Awarding Institution	The Hong Kong University of Science and Technology
Supervisor	Huamin QU (Supervisor) & David Kei Man YIP (Supervisor)