[Paper Review] Mapping Tasks to Interactions for Graph Exploration and Graph Editing on Interactive Surfaces

Stefan Gladisch, Ulrike Kister+3 more|arXiv (Cornell University)|Apr 29, 2015

Usability and User Interface Design25 references4 citations

TL;DR

This paper proposes a systematic framework for mapping graph exploration and editing tasks to interactions on interactive surfaces like tablets and touchscreens. It identifies core tasks, analyzes interaction modalities (touch, pen, tangible), and introduces a formal optimization approach using weighted criteria to generate high-quality, reusable interaction mappings—offering a reusable guideline for designing intuitive, unified graph visualization systems for modern devices.

ABSTRACT

Graph exploration and editing are still mostly considered independently and systems to work with are not designed for todays interactive surfaces like smartphones, tablets or tabletops. When developing a system for those modern devices that supports both graph exploration and graph editing, it is necessary to 1) identify what basic tasks need to be supported, 2) what interactions can be used, and 3) how to map these tasks and interactions. This technical report provides a list of basic interaction tasks for graph exploration and editing as a result of an extensive system review. Moreover, different interaction modalities of interactive surfaces are reviewed according to their interaction vocabulary and further degrees of freedom that can be used to make interactions distinguishable are discussed. Beyond the scope of graph exploration and editing, we provide an approach for finding and evaluating a mapping from tasks to interactions, that is generally applicable. Thus, this work acts as a guideline for developing a system for graph exploration and editing that is specifically designed for interactive surfaces.

Motivation & Objective

• Address the lack of unified systems that support both graph exploration and editing on modern interactive surfaces like tablets and touchscreens.
• Identify and classify fundamental interaction tasks required for graph exploration and editing through a comprehensive review of existing systems.
• Analyze interaction modalities (touch, pen, tangible) and their vocabularies, including degrees of freedom to distinguish gestures.
• Develop a formal, generalizable method for mapping tasks to interactions that ensures usability and consistency.
• Provide a reusable design guideline for creating interactive graph visualization systems tailored to interactive surfaces.

Proposed method

• Conduct a systematic review of 12 graph exploration and editing systems (e.g., Gephi, Cytoscape, Tulip) to extract and classify core interaction tasks.
• Analyze interaction modalities—touch, pen, and tangible—by reviewing their interaction vocabularies and degrees of freedom (e.g., pressure, duration, spatial configuration).
• Define a formal mapping function $ m: T \rightarrow I $, where $ T $ is the set of tasks and $ I $ is the set of interactions, ensuring injectivity to avoid ambiguity.
• Establish a set of interaction criteria (e.g., predictability, consistency, familiarity, viscosity, recoverability, directness, continuity) as design guidelines.
• Formalize the quality of a mapping using a weighted quality function $ \hat{q} = \frac{\sum_{i=1}^{n} \alpha_i \cdot q(m, c_i)}{n} $, where $ \alpha_i $ are weights and $ q $ measures adherence to each criterion.
• Frame the mapping process as an optimization problem to maximize overall quality, enabling systematic evaluation and selection of optimal task-interaction mappings.

Experimental results

Research questions

• RQ1 What fundamental tasks are required for effective graph exploration and editing on interactive surfaces?
• RQ2 Which interaction modalities (touch, pen, tangible) and their associated gestures are most suitable for graph manipulation?
• RQ3 How can a systematic and reusable method be developed to map graph tasks to appropriate interactions?
• RQ4 What criteria ensure that a task-interaction mapping is usable, consistent, and intuitive across different contexts?
• RQ5 How can the quality of a task-interaction mapping be formally evaluated and optimized?

Key findings

• The authors identified and classified a comprehensive set of basic interaction tasks essential for graph exploration and editing, derived from a review of 12 existing systems.
• The study reveals that interaction modalities on interactive surfaces offer distinct vocabularies and degrees of freedom (e.g., pressure, duration, spatial gestures) that can be leveraged to create distinguishable and expressive interactions.
• The formal mapping approach using weighted quality functions enables systematic evaluation of mappings based on criteria like predictability, consistency, and viscosity.
• The optimization framework allows for the identification of high-quality mappings by maximizing a normalized, weighted quality score across multiple design criteria.
• The proposed method is generalizable beyond graph visualization and can be applied to any system requiring task-interaction mapping on interactive surfaces.
• Despite the formal framework, the full mapping for all tasks and modalities remains computationally intensive and is left for future work.