Understanding the Temporal Rhythm of User Engagement in Online Creative Community with Data-Driven Approaches

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Understanding the Temporal Rhythm of User Engagement in Online Creative 
Community with Data-Driven Approaches"

By

Mr. Qingyu GUO


Abstract:

Online creative communities (OCCs) facilitate the exchange of ideas, 
perspectives, and cultural influences among creators and fans, enriching the 
creative landscape with diverse voices. Encouraging user engagement in OCCs 
(e.g., community participation) is essential for building a thriving creative 
ecosystem.

Recognizing the temporal nuances of user behaviors and engagement in online 
communities is crucial for stakeholders to enhance community design and user 
management. However, the temporal aspect of user engagement in creative 
contexts remains underexplored. This dissertation addresses this gap by 
analyzing user engagement under representative temporal rhythms, including 
procedural rhythm (associated with routine creative processes), transient 
episodic rhythm (shaped by short-term events), and extended episodic rhythm 
(influenced by long-term events). We gathered large-scale user interaction data 
from OCCs and utilized data- driven approaches to investigate the issue.

The dissertation starts with art learning communities where creators seek and 
provide critiques for original artworks. The creative process presents a 
procedural rhythm, as creators could seek feedback in various creation stages 
(i.e., work-in-progress and complete work). We quantitatively examine how 
creators' engagement and expected feedback are affected by such procedural 
rhythm. We then delve into online music fandoms, where fans create, collect, 
and discuss the content of their mutual interest (e.g., music artists). 
Transient episodic rhythms emerge in fandom communities as celebrity events 
could frequently attract newcomers. We quantitatively analyze user behaviors 
and their correlations with users' future engagement in the community across 
the transient episodic rhythm. Finally, we examine a leading art-sharing 
community, Deviantart, where creators share and comment on artworks. Recent 
innovations in generative AI are transforming user-generated artworks into 
AI-generated artworks, leading to an extended episodic rhythm in the creative 
community. We uncover shifts in community interaction since the enduring 
episodic rhythm to understand the evolvement of community members' practice and 
motivation in engaging with the community.

Collectively, this dissertation highlights the significance of understanding 
user engagement in OCCs through the lens of temporal rhythms. The findings 
offer valuable insights and design implications for cultivating user engagement 
in dynamic online creative environments and fostering a sustainable creative 
ecosystem.


Date:                   Tuesday, 13 August 2024

Time:                   10:00am - 12:00noon

Venue:                  Room 5501
                        Lifts 25/26

Chairman:               Dr. Xiaowei ZHANG (IEDA)

Committee Members:      Dr. Xiaojuan MA (Supervisor)
                        Dr. Qifeng CHEN
                        Prof. Andrew HORNER
                        Prof. Fugee TSUNG (IEDA)
                        Dr. Kening ZHU (CityU)