Sparklis

May 08, 2017 by Sébastien Ferré, Categories: User Applications, Query Builders, SPARQL query interfaces

Sparklis is a query builder in natural language that allows people to explore and query SPARQL endpoints with all the power of SPARQL and without any knowledge of SPARQL.

Sparklis is a Web client running entirely in the browser. It directly connects to SPARQL endpoints to retrieve query results and suggested query elements. It covers a large subset of SPARQL 1.1 SELECT queries: basic graph patterns including cycles, UNION, OPTIONAL, NOT EXISTS, FILTER, BIND, complex expressions, aggregations, GROUP BY, ORDER BY. All those features can be combined in a flexible way, like in SPARQL. Results are presented as tables, and also on maps. A configuration panel offers a few configuration options to adapt to different endpoints (e.g., GET/POST, labelling properties and language tags). Sparklis also includes the YASGUI editor to let advanced users access and modify the SPARQL translation of the query.

On the "Examples" page there are 100+ example queries over several datasets, and a number of them have a YouTube screencast to show how they are built.

Sparklis reconciles expressivity and usability in semantic search by tightly combining a Query Builder, a Natural Language Interface, and a Faceted Search system. As a Query Builder it lets users build complex queries by composing elementary queries in an incremental fashion. An elementary query can be a class (e.g., "a film"), a property (e.g., "that has a director"), a RDF node (e.g., "Tim Burton"), a reference to another node (e.g., "the film"), or an operator (e.g., "not", "or", "highest-to-lowest", "+", "average").

As a Faceted Search system, at every step, the query under construction is well-formed, query results are computed and displayed, and the suggested query elements are derived from actual data - not only from schema - so as to prevent the construction of non-sensical or empty results. The display of results and data-relevant suggestions at every step provides constant and acurate feedback to users during the construction process. This supports exploratory search, serendipity, and confidence about final results.

As a Natural Language Interface, everything presented to users - queries, suggested query elements, and results - are verbalized in natural language, completely hidding SPARQL behind the user interface. Compared to Query Answering (QA) systems, the hard problem of spontaneous NL understanding is avoided by controlling query formulation through guided query construction, and replaced by the simpler problem of NL generation. The user interface lends itself to multilinguality, and is so far available in English, French, and Spanish.


Online tool: http://www.irisa.fr/LIS/ferre/sparklis/
 

Team: SemLIS at IRISA http://www-semlis.irisa.fr/
Contact: Sébastien Ferré http://people.irisa.fr/Sebastien.Ferre/