Reverse Engineering SPARQL Queries

Abstract

Semantic Web systems provide open interfaces for end-users to access data via a powerful high-level query language, SPARQL. But users unfamiliar with either the details of SPARQL or properties of the target dataset may find it easier to query by example-give examples of the information they want (or examples of both what they want and what they do not want) and let the system reverse engineer the desired query from the examples. This approach has been heavily used in the setting of relational databases. We provide here an investigation of the reverse engineering problem in the context of SPARQL. We first provide a theoretical study, formalising variants of the reverse engineering problem and giving tight bounds on its complexity. We next explain an implementation of a reverse engineering tool for positive examples. An experimental analysis of the tool shows that it scales well in the data size, number of examples, and in the size of the smallest query that fits the data. We also give evidence that reverse engineering tools can provide benefits on real-life datasets. * The authors would like to thank Michael Benedikt for many fruitful discussions about the results presented in this paper. M. Arenas was funded by Millennium Nucleus Center for Semantic Web Research under Grant NC120004, and G. I. Diaz by Becas Chile of CONICYT Chile. Copyright is held by the International World Wide Web Conference Committee (IW3C2). IW3C2 reserves the right to provide a hyperlink to the author's site if the Material is used in electronic media.

Ontology-Mediated Querying with Horn Description Logics

“…This problem is related to machine learning research: We want to learn a query from the given examples. QBE has been suggested in [30] and has been studied for traditional databases and different query languages [1,5,8,9,13,27]. We initiate the research on QBE for OMQs.…”

New trends on exploratory methods for data analytics

“…For graph data there are two prominent approaches: the first use subgraphs, or partially specified structures as input examples [1,7,13,19], while the second focuses on the vertices of the graph, which are used for making the selections [2,14,21]. Structures convey a more precise information and therefore can be used to quickly prune the search space.…”

“…Among the existing approaches exemplar queries [18,19] and Graph Query by Example (GQBE) [13] use subgraph isomorphism or structural similarities to identify structures related to the one the user provided. A different approach is the reverse engineering of SPARQL queries [1,7] in which the input is a set of positive and negative entity mentions in a RDF dataset. This approach is similar to those discussed for the relational case, and is related to learning path or join queries given positive and negative nodes [2,3].…”

“…-Similar entity search [28] -Structure-based approaches: Exemplar Queries [19], Graph Query By Example [13] -Learning Path queries [2,3] -Reverse engineering of SPARQL queries [1,7] -Node-based approaches: to discover communities [14], dense regions in the graph [10,24], and outliers [21].…”

Ontology-Mediated Queries from Examples: a Glimpse at the DL-Lite Case

“…Since it was formulated by Zloof in the 70s [23], researchers have come across it in data exploration and analysis, usability, data security, the study of expressiveness of query languages, etc., see [15,16]. It has been studied for different query languages and different types of databases, including relational data [5,19,21,14,20], graph databases [7,1], and RDF data [2]. Much work has aimed at finding practicable solutions [15,20,14,11,9].…”