Vocabulary for annotating questions with information generated in Question Answering piplines
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
README.md
qanary.owl
qanary.ttl

README.md

QAOntology

This repository contains the Qanary (Question answering vocabulary) vocabulary. The permanent URI of the ontology is:

https://w3id.org/wdaqua/qanary#

The Qanary vocabulary aims to define a vocabulary to express information that are generated during the execution of a Question Answering (QA) pipeline on a question given as speech or text input.

Consider for example the question "When was Barack Obama born?". Typical information generated by components of a QA pipeline are the position of named entities (NE) (like "Barack Obama"), the relations of the ontology used to express the relational phrase in the question (that "born" refers to dbo:birthPlace), the expected answer (here: a date), and the generated SPARQL query. Other such information is: the language of the question, possible ontologies that can be used to answer it or if a question is a text or audio file.

The Qanary vocabulary was introduced for the first time in:

Towards a Message-Driven Vocabulary for Promoting the Interoperability of Question Answering Systems, Kuldeep Singh and Andreas Both and Dennis Diefenbach and Saeedeh Shekarpour, Proc. of the 10th IEEE Int. Conf. on Semantic Computing (ICSC) 2016.