Origraph: Interactive Network Wrangling

Abstract

Figure 1: Overview of the Origraph UI. The network model view shows relationships between node and edge classes and is the primary interface for operations related to connectivity. The attribute view shows node and edge attributes in a table and is the primary interface for attribute-related operations. The network sample view visualizes a preview of the current state of the network. ABSTRACTNetworks are a natural way of thinking about many datasets. The data on which a network is based, however, is rarely collected in a form that suits the analysis process, making it necessary to create and reshape networks. Data wrangling is widely acknowledged to be a critical part of the data analysis pipeline, yet interactive network wrangling has received little attention in the visualization research community. In this paper, we discuss a set of operations that are important for wrangling network datasets and introduce a visual data wrangling tool, Origraph, that enables analysts to apply these operations to their datasets. Key operations include creating a network from source data such as tables, reshaping a network by introducing new node or edge classes, filtering nodes or edges, and deriving new node or edge attributes. Our tool, Origraph, enables analysts to execute these operations with little to no programming, and to immediately visualize the results. Origraph provides views to investigate the network model, a sample of the network, and node and edge attributes. In addition, we introduce interfaces designed to aid analysts in specifying arguments for sensible network wrangling operations. We demonstrate the usefulness of Origraph in two Use Cases: first, we investigate gender bias in the film industry, and then the influence of money on the political support for the war in Yemen. [Graph-based database models]: -the way an analyst thinks about it. To model data as a network, analysts must wrangle the dataset, often starting with tabular or key-value data. Transforming data itself can lead to new hypotheses, and thus a new network representation of the data. Also, new tasks often necessitate new data abstractions [40]. It stands to reason that the ability to rapidly and easily transform network data can foster creative visualization solutions and simplify both exploration and communication of the key aspects of a dataset. Existing network wrangling tools, most notably Ploceus and Orion [21,33], focus on creating an initial network model, but no tools yet exist to iteratively and interactively reshape the network model itself with operations such as converting between nodes and edges [41]. Other operations that leverage edges, such as arXiv:1812.06337v3 [cs.HC]

The State of the Art in Visualizing Multivariate Networks

“…However, many other operations are conceivable. For a more comprehensive discussion of network wrangling operations, refer to the work by Bigelow et al [BNML18].…”

“…Hence, it should be applied with care. We believe that it is most useful as a preprocessing operation in dedicated graph wrangling tools [BNML18].…”

Exploring the Personal Informatics Analysis Gap: "There's a Lot of Bacon"

“…The results of these studies inform a growing ecosystem of tools for data wrangling [51,52], interactive visual analysis [53,54], and visualization recommendations [55,56,57].…”

Taggle: Combining overview and details in tabular data visualizations

“…Examples of this approach are Ploceus, 14 Orion, 15 and Origraph. 16 We argue that projection techniques are well suited to visualize structure in a high-dimensional dataset, but they cannot adequately show why items in a cluster belong together. Projection techniques are especially useful in cases where the dimensions themselves are not meaningful to human analysts, such as a table of term frequencies when analyzing text documents.…”