QA3: A natural language approach to question answering over RDF data cubes

Issue title: Special Issue on Expressive Natural Language Interfaces for the Semantic Web

Guest editors: Stamatia Dasiopoulou, Georgios Meditskos, Leo Wanner, Stefanos Vrochidis and Philipp Cimiano

Article type: Research Article

Authors: Atzori, Maurizio^{a; *} | Mazzeo, Giuseppe M.^b | Zaniolo, Carlo^b

Affiliations: [a] Department of Maths and Computer Science, University of Cagliari, V. Ospedale 72, 09124 Cagliari, Italy. E-mail: atzori@unica.it | [b] Computer Science Department, UCLA, 3532D Boelter Hall Los Angeles, CA 90095-1596, CA, USA. E-mails: mazzeo@cs.ucla.edu, zaniolo@cs.ucla.edu

Correspondence: [*] Corresponding author. E-mail: atzori@unica.it.

Abstract: In this paper we present QA3, a question answering (QA) system over RDF data cubes. The system first tags chunks of text with elements of the knowledge base, and then leverages the well-defined structure of data cubes to create a SPARQL query from the tags. For each class of questions with the same structure a SPARQL template is defined, to be filled in with SPARQL fragments obtained by the interpretation of the question. The correct template is chosen by using an original set of regex-like patterns, based on both syntactical and semantic features of the tokens extracted from the question. Preliminary results obtained using a limited set of templates are encouraging and suggest a number of improvements. QA3 can currently provide a correct answer to 27 of the 50 questions of the test set of the task 3 of QALD-6 challenge, remarkably improving the state of the art in natural language question answering over data cubes.

Keywords: Question answering, RDF data cube, statistical queries, free natural language

DOI: 10.3233/SW-180328

Journal: Semantic Web, vol. 10, no. 3, pp. 587-604, 2019