Survey on the Analysis of User Interactions and Visualization Provenance

Abstract

There is fast-growing literature on provenance-related research, covering aspects such as its theoretical framework, use cases, and techniques for capturing, visualizing, and analyzing provenance data. As a result, there is an increasing need to identify and taxonomize the existing scholarship. Such an organization of the research landscape will provide a complete picture of the current state of inquiry and identify knowledge gaps or possible avenues for further investigation. In this STAR, we aim to produce a comprehensive survey of work in the data visualization and visual analytics field that focus on the analysis of user interaction and provenance data. We structure our survey around three primary questions: (1) WHY analyze provenance data, (2) WHAT provenance data to encode and how to encode it, and (3) HOW to analyze provenance data. A concluding discussion provides evidence-based guidelines and highlights concrete opportunities for future development in this emerging area.

Towards a Visual Approach for Representing Analytical Provenance in Exploration Processes

“…There are several reasons (why) for provenance analysis, such as understanding users, evaluating systems and algorithms, building adaptive systems, model steering, replicating, verifying and re-analyzing, reporting and storytelling [13]. Our work aims to allow users to recreate their analytical reasoning process, while supporting verification, replication, or reapplication of analysis sessions.…”

“…The second aspect (what) refers to the way provenance data is represented. Among the schemes of provenance data representation (e.g., grammars, models, graphs), the sequencebased scheme is the most common [13], and represents the user's interactions as a list of actions. Graphs are used to connect entities (or concepts) that change state during the exploration process, or temporal events, as in history graphs (e.g., VisTrails, GraphTrail, and SOMFlow).…”

“…Finally, as for the how aspect, previous works have addressed provenance analysis using methods such as classification and statistical modeling techniques, pattern analysis, probabilistic techniques for prediction, grammar and scripts analysis, and interactive visual analysis [13]. Although this paper do not target the analysis of provenance data, the explicit representation of provenance information supported by our approach allows anyone to perform an interactive visual analysis on the exploratory flows by querying the parameters used in each step of the process.…”

“…Moreover, the validation of hypotheses might require branching out the exploratory path to compare data observed in different views. Evidence of these exploratory processes has promoted rapid growth in research on analytical provenance [12] [13], including techniques to capture, visualize, and analyze provenance information. Most visualization systems record the user's actions through history graphs [14], which is often not enough to analyze the analytical process [15].…”

Reusing Interactive Analysis Workflows

“…An alternative approach to explicit workflow modeling is tracking the interaction provenance [26,27] and using this information to later extract workflows. Although several visualization systems track provenance [28,29,30,31] and a few dedicated libraries to track provenance exist [32,33], most tools do not explicitly curate workflows based on the provenance.…”