The hackathon will be held virtually from 5-9 July, 2021
Many question answering systems over RDF that are induced from pairs of questions and answers using some machine learning technique suffer from a lack of controllability, making the governance and incremental improvement of the system challenging, not to mention the initial effort of collecting and providing training data. As an alternative, we present a model-based approach to question answering that uses an ontology lexicon in lemon format and automatically generates a lexicalized grammar that can be used to interpret and parse questions into SPARQL queries. The approach gives maximum control over the QA system to the developer of the system as every lexicon entry added to the lexicon increases the coverage of the grammar, and thus of the QA system, in a predictable way.
The hackathon aims to improve the QA system on the combination of semantic web and language technologies. There are several possibilities to extend the system such as:
These are to be understood as suggestions from our side. We are also open to other directions or tasks participants are interested to work on.
The hackathon is supported by the project Pret-A-LLOD as well as by Nexus Linguarum Cost Action. We welcome cross-disciplinary collaborations of computer scientists, computational linguistics and related fields to participate and contribute any level of their interest. Please register the hackathon for participation until 28 June 2021
Philipp Cimiano (CITEC - University Bielefeld, Germany), ChairMohammad Fazleh Elahi (CITEC - University Bielefeld, Germany)Frank Grimm (CITEC - University Bielefeld, Germany)Basil Ell (CITEC - University Bielefeld, Germany and Department of Informatics, University of Oslo, Norway)
1.Automatic Grammar Generation: Benz, V., Cimiano, P., Elahi, M.F., Ell, B. (2020), Generating Grammars from lemon lexica for Questions Answering over Linked Data: a Preliminary Analysis. In: NLIWOD workshop at ISWC. vol. 2722, pp. 40–55. CEUR-WS.org (2020).
2. Question Answering system: Elahi, M.F, Ell, B, Cimiano, P.(2021) Question Answering over RDF Data based on Grammars Automatically Generated from Lemon Models
Philipp Cimiano, Paul Buitelaar, John McCrae, Michael Sintek. LexInfo: A Declarative Model for the Lexicon-Ontology Interface
Ricardo Usbeck, Axel-Cyrille Ngonga Ngomo, Bastian Haarmann,Anastasia Krithara, Michael R¨oder, and Giulio Napolitano. 7th Open Challenge on Question Answering over Linked Data (QALD-7) .
John P. McCrae, Francesca Quattri, Christina Unger, Philipp Cimiano. Modelling the Semantics of Adjectives in the Ontology-Lexicon Interface. Proceedings of the 4th Workshop on Cognitive Aspects of the Lexicon CogALex(2014).
Mohammad Fazleh Elahi
e-mail: melahi[at sign]techfak.uni-bielefeld.de