Creating quranic question taxonomy

Author

Saeedi, Parvaneh ; Heidari, Samaneh ; Farhoodi, Mojgan

Author_Institution

Inf. Technol. Fac., Iran Telecommun. Res. Center, Tehran, Iran

fYear

2014

fDate

20-22 May 2014

Firstpage

1070

Lastpage

1074

Abstract

The knowledge of answer type is a valuable information in question answering(QA) systems which can be applied in searching for the answers. In this paper, we have prepared a question taxonomy which cover answer types in Quranic domain. It has 83 fine-grained and 30 coarse-grained semantic classes which are in a hierarchical structure. These semantic classes are extracted from about 6000 Quranic questions which are gathered from credible websites. This taxonomy is evaluated using the Quranic ontology which is currently developed in Quranjooy¹ project at ITRC². Quranic taxonomy has been applied in question classification and answer extraction of Quranjooy. it can significantly improve the answer extraction performance.

Keywords

Web sites; ontologies (artificial intelligence); question answering (information retrieval); ITRC; QA; Quranic ontology; Quranic question taxonomy; Quranic questions; Quranjooy project; answer extraction performance; coarse-grained semantic classes; credible websites; fine-grained semantic classes; question answering systems; Data mining; Knowledge discovery; Ontologies; Semantics; Speech; Taxonomy; Uniform resource locators;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical Engineering (ICEE), 2014 22nd Iranian Conference on

Conference_Location

Tehran

Type

conf

DOI

10.1109/IranianCEE.2014.6999694

Filename

6999694