Beast to beauty transition: Design process for complex visual analytics

Abstract

Visual analytics fundamentally influences the analytical process. Without proper guidance, the complex condition can become a beast that hides the beauty of analytical data. Since visual analytics is the only way to represent analytical outcomes, ineffective visual analytics will bury the relevancy of information to support any valuable decision. Thus, this research revisits the way to design more valuable and effective visual analytics, especially for the complex condition which involves i) the information complexities, ii) the nature of the activities involved and iii) the complex context of usage. Governed by the Design Science Research Methodology (DSRM), the research applied the pragmatic philosophical worldview and exploratory research. Furthermore, the design process embedded the human-activity centered design approaches as the new perspective to get a better understanding of the visual analytics users and the analytical activities involved. There are basically three important activities of the design process that comprises of five design phases. Activity 1 aims to identify the visual analytics context and its challenges, Activity 2 to develop the solution of visual analytics design and Activity 3 to evaluate the effectiveness of visual analytics to handle its complex context. By offering a set of design process, this research aims to improve the experience of the visual analytics community to design in a more flexible, dynamic and intuitive way. Finally, this paper also recommends appropriate methods, the unit of analysis, sampling strategy and data management & analysis suitable for complex visual analytics.