Despite the growing popularity of knowledge graphs for managing diverse data at large scale, users who wish to pose expressive queries against such graphs are often expected to know (i) how to formulate queries in a language such as SPARQL, and (ii) how entities of interest are described in the graph. In this demo we present a system that allows non-expert users to simultaneously navigate and query knowledge graphs by incrementally exploring the dataset.
In a previous work [
Most popular user interfaces to query SPARQL endpoints simply propose a
text box that allows end users to write a query. However, using such interface
for querying graphs is challenging. First, users are required to have technical
knowledge of the query language and the semantics of its operators. Second,
graph data usually represents a variety of data from di erent domains, meaning
that the users may not be easily able to conceptualize the data that they are
querying. Despite these limitations, query services for DBpedia and Wikidata
are receiving in the order of millions of queries per day [
Some approaches adopted by interfaces that allow lay users to visualise, search, browse and query knowledge graphs involve keyword search, faceted browsing, graph-based browsing, query building, graph summarization, visualization techniques, and combinations thereof. Many of these systems however, foster usability by removing SPARQL features such as the use of the OPTIONAL operator or cycles within the query. Furthermore, interfaces that allow cycles within graph patterns assume technical expertise of the query language and/or knowledge of how data are modeled.
Here we present an interface that aims to allow lay users to build and execute graph-pattern queries on knowledge graphs; within our proposal, the user navigates a visual representation of a sub-graph, and in so doing, incrementally builds a potentially complex (cyclical) graph pattern. The interface can be accessed at https://www.rdfexplorer.org and the code can be found in its corresponding GitHub repository4. 2
In this section, we present the RDF Explorer system, whose goal is to enable lay
users to query and explore RDF graphs. We present the main characteristics of
the system as well as a description of how the overall system is implemented.
RDF Explorer Interface The interface is composed of six main components
displayed in three panes. Figure 1 provides a screenshot of the interface for
querying Wikidata, where we can see three components: a search panel (left
pane), a visual query editor (center pane), a node detail view (right pane); in the
top right corner are buttons to switch the right pane to display one of the three
other components: a node editor (allowing to add restrictions to a highlighted
node), a SPARQL query editor (showing the current query), and a help panel.
{ The search panel allows the user to search for an RDF resource (in Wikidata
currently, Figure 1 left). This feature uses a Wikidata index [
A Visual Query Builder for Semantic Web Knowledge Graphs Query creation process The process starts with a blank visual query editor. The user must then start by adding a new node, be it a variable node or a constant node; for selecting x, the user can type a keyword phrase into the search pane on the left, which will generate autosuggestions, where any of the results shown can be dragged into the central query editor pane. The user may then proceed to add a second node by the same means. With two or more nodes available, the user can now click and drag between two nodes to generate an edge with a variable edge-label (shown as a box nested inside the source node); a list of potential IRIs will be suggested for replacing the variable, where only IRIs that generate non-empty results for the underlying query will be o ered.
In this demo we presented the companion user interface to the paper in [