PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Contextualized question answering
Luka Bradeško, Lorand Dali, Blaz Fortuna, Marko Grobelnik, Dunja Mladenić, Inna Novalija and Boštjan Pajntar
In: ITI 2011, 27-30 Jun 2011, Cavtat, Dubrovnik, Croatia.

Abstract

The paper proposes an approach to contextualized answering of questions. The contextualization is achieved by using an ontology. The answers are provided based on a domain specific document collection. The approach consists of several phases as follows: data preparation, data enhancement, data indexing and handling questions. The functioning of the proposed approach is demonstrated on English document collections on Aquatic Sciences and Fisheries - ASFA, using Cyc ontology, ASFA thesaurus as domain specific ontology and WordNet as general ontology. The approach is in general applicable to other datasets in other languages, assuming the necessary natural language processing is handled and the match between the document collection language, domain ontology language and Cyc ontology is ensured.

EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Information Retrieval & Textual Information Access
ID Code:8721
Deposited By:Jan Rupnik
Deposited On:21 February 2012