Arabic collocations extraction using Gate

Author

Zaidi, Soraya ; Laskri, M.T. ; Abdelali, Ahmed

Author_Institution

Dept of Comput. Sci., Badji Mokhtar Univ., Annaba, Algeria

fYear

2010

fDate

3-5 Oct. 2010

Firstpage

473

Lastpage

475

Abstract

Information extraction (IE) from corpora is texts analysis in order to extract structured information such as Named Entities (NE) which may be names of person, organization, address, date, location etc. ... GATE is a software toolkit written in Java from 1995 and widely used worldwide by many communities (scientists, companies, teachers, students) for natural language processing. We have experimented Gate for extracting terms by writing new Jape rules (Java Annotation Pattern Engine) and used them on a tagged corpus developed at Leeds University. These terms will be used in the texts-based ontologies building. In our case this ontology will be incorporated into a search engine to expand queries on the Web, in the specified domain.

Keywords

Java; natural language processing; ontologies (artificial intelligence); query processing; search engines; software tools; text analysis; Arabic collocation; GATE; Jape rules; Java; Java annotation pattern engine; information extraction; named entity; natural language processing; ontology; search engine; software toolkit; tagged corpus; texts analysis; Computer architecture; Data mining; Grammar; Logic gates; Ontologies; Software; Transducers; Collocation extraction; GATE; JAPE; NLP; Ontologies; Textual engineering;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine and Web Intelligence (ICMWI), 2010 International Conference on

Conference_Location

Algiers

Print_ISBN

978-1-4244-8608-3

Type

conf

DOI

10.1109/ICMWI.2010.5648038

Filename

5648038