Propagation of change and visualization of causality in dependency structures

Analysts use visual analytics tools to gain insight through data. But one problem of these tools is the complexity of the analytics process itself. Sometimes the analyst has to repeat the same task during the analysis; to rerun some previous queries or to update the state of the system by the arrival of new data. CZSaw is a visual analytics tool that stores the process of analysis and visualizes it. The dependency graph of CZSaw allows parts of the analytical process to be reused on new data to prevent repetition of those parts. This document describes the design and development of CZSaw’s dependency graph as well as its visualization. A major challenge is visualization of causality in the graph when an update gets propagated. We compare three different causality visualization methods in terms of task completion duration and accuracy in a study.