ZipLine: Visual Analysis of Multivariate Graphs with Predicate Logic
Authors
Sjoerd Vink (Utrecht University), Susie S.Y. Li (Tufts University), Brian Montambault (Tufts University), Michael Behrisch (Utrecht University), Mingwei S.G. Li (Tufts University), Remco Chang (Tufts University)
Presentation
- Session
- I feel tangled in a net
- Time
- Wednesday, Nov 11, 08:00 – 08:12 (US/Eastern) · session 08:00 – 09:30
- Location
- Hall America center
Links
Sign in to access the preprint PDF.
Sign in- Download Supplemental Material
Keywords
Multivariate graph analytics, First-order logic, Predicate induction, Graph visualization
Abstract
Multivariate graphs unite two distinct data perspectives: a topological structure defined by nodes and edges, and attribute data associated with each node. Analyzing such graphs therefore requires reasoning across two complementary spaces. However, existing systems typically emphasize the analysis of one space at a time, focusing either on topology or on attributes. As a result, exploration, analysis, and pattern discovery that depend on their interaction remain difficult. In this paper, we present a technique for formalizing and inducing patterns in multivariate graphs, instantiated in the ZipLine visual analytics system. Its central contribution is a predicate formalism that represents structural properties, node attributes, and bounded neighborhood relations while ensuring deterministic evaluation and a finite search space. Building on this formalism, ZipLine provides a predicate-learning algorithm that induces logical expressions from analyst-defined selections across topology and attribute views by combining attribute filters, structural properties, and neighborhood criteria. This approach supports iterative analysis by enabling analysts to refine patterns through coordinated reasoning over topology and attributes. We demonstrate ZipLine through three case studies in energy infrastructure, cybersecurity, and drug discovery analysis. The results show that ZipLine enables expressive multivariate graph analysis through unified reasoning across topology and attributes.
For Practitioners
This paper is relevant to data scientists, graph analysts, visual analytics researchers, and domain practitioners working with networked data, including energy infrastructure planners, cybersecurity analysts, and biomedical researchers. Practitioners can apply the predicate formalism to express patterns that combine node attributes, structural properties, and neighborhood relations. They can also use the predicate-learning approach to translate visually selected subgraphs into explicit, inspectable, and reusable logical expressions for exploration, comparison, and hypothesis generation. System developers can incorporate these techniques into graph-analysis tools to connect interactive visualization with formal querying and pattern induction.