FACETS: Adaptive Local Exploration of Large Graphs

Abstract

Visualization is a powerful paradigm for exploratory data analysis. Visualizing large graphs, however, often results in excessive edges crossings and overlapping nodes. We propose a new scalable approach called FACETS that helps users adaptively explore large million-node graphs from a local perspective, guiding them to focus on nodes and neighborhoods that are most subjectively interesting to users. We contribute novel ideas to measure this interestingness in terms of how surprising a neighborhood is given the background distribution, as well as how well it matches what the user has chosen to explore. FACETS uses Jensen-Shannon divergence over information-theoretically optimized histograms to calculate the subjective user interest and surprise scores. Participants in a user study found FACETS easy to use, easy to learn, and exciting to use. Empirical runtime analyses demonstrated FACETS's practical scalability on large real-world graphs with up to 5 million edges, returning results in fewer than 1.5 seconds.

Using knowledge anchors to facilitate user exploration of data graphs

“…used by the CAC value, the CU will also include the proportion of all triples between the subject entity u3 and all the object entities in the graph (i.e. entities 12 v , 21 v , 22 v ) linked via relationship D (i.e. 3 triples) over the number of entities linked via subsumption relationships (e.g.…”

“…novices in the domain. Examples include: personalising the exploration path tailored to the user's interests [10], presenting RDF patterns to give an overview of the domain [11], or providing graph visualisations to support navigation [12]. However, existing work on facilitating users' exploration through data graphs has addressed mainly investigative tasks, omitting important exploratory search tasks linked to supporting learning.…”

PERSEUS-HUB: Interactive and Collective Exploration of Large-Scale Graphs

“…Refs. [7,13,14] take an exploratory search approach based on the assumption that users are non-experts and unfamiliar with their datasets. Therefore, they guide users to perform analysis by incrementally querying information or browsing through the various relationships in the datasets to discover any unexpected results.…”

“…Therefore, they guide users to perform analysis by incrementally querying information or browsing through the various relationships in the datasets to discover any unexpected results. FACETS [13] allows users to adaptively explore large graphs by only showing their most interesting parts. The interestingness of a node is based on how surprising its neighbor's data distributions are, as well as how well it matches what the user has chosen to explore.…”

“…GLO-STIX [7] ? FACETS [13] Voyager 2 [14] Signposts [15] NodeTrix [16] PivotSlice [17] Refinery [18] Apolo [8] OPAvion [19] NetRay [22] Motif-Simp [23] PERSEUS [9,10] PERSEUS-HUB…”