Teng, Z. (CM) – Visualizing Player Processes: Towards Design Guidelines for Interactive Process Visualization Tools in Game Analytics

Game analysts face a significant challenge in understanding problem-solving and decision-making processes from the vast and complex sequential data generated by modern video games. Existing visualization tools often fail to adequately support the exploration, suffering from issues of visual clutter, inflexible cohort construction, and a lack of interactive depth. To address this gap, this dissertation adopts a Research through Design (RtD) methodology to investigate how an interactive process visualization system can be designed and developed to better support the needs of game analysts.

The research was conducted in three phases. First, an initial set of five design guidelines was identified through a breakdown analysis of existing tools and semi-structured interviews with professional game analysts. Second, these guidelines were iteratively refined through long-term, collaborative case studies with analysts working on diverse commercial games. This process not only validated the initial guidelines and surfaced one additional guideline concerning interactive inspection, but also resulted in INSPECT, an interactive process visualization prototyping system that embodies the refined guidelines. Third, the guidelines were empirically validated through two complementary user studies of the INSPECT system. A controlled experiment demonstrated that features designed according to the guidelines enabled participants to identify player strategies more efficiently than with a baseline system, while a qualitative study with professional Dota 2 coaches and players demonstrated the system’s practical value for strategic analysis and strong usability.

The primary contributions of this dissertation to the fields of game analytics and information visualization is a set of validated design guidelines for process visualization tools. This contribution provides a durable and transferable framework for the design and development of more effective, analyst-centered tools for understanding player problem-solving and decision-making processes.

Event Host: Zhaoqing Teng, Ph.D. Candidate, Computational Media 

Advsior: Magy Seif El-Nasr

Zoom- https://ucsc.zoom.us/j/97624383966?pwd=NGolaaTbhdytPcDK6aRIBDIv63b8lm.1

Passcode-595285