Semantic Annotation of Arabic Web Documents using Deep Learning

Saeed Albukhitan, Ahmed Alnazer, Tarek Helmy*

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

7 Scopus citations

Abstract

The vision of Semantic Web is to have a Web of things instead of Web of documents in a form that can be processed by machines. This vision could be achieved on the existing Web using semantic annotation based on common and public ontologies. Due to exponential growth and the huge size of the Web sources, there is a need to have fast and automatic semantic annotation of Web documents. Arabic language received less attention in semantic Web research as compared to Latin languages especially in the field of semantic annotation. The aim of this paper is to investigate the feasibility of using word embeddings from deep learning algorithms for semantic annotation of Arabic Web documents. To evaluate the performance of the proposed framework, food, nutrition, and health ontologies were used to annotate some related Web documents. For a given set of Arabic documents and ontologies, the framework produces annotations of these documents using different output formats. The initial results show a promising performance which will support the research in the Semantic Web with respect to Arabic language. The proposed framework could be used for building semantic Web application and semantic search engines for Arabic Language.

Original languageEnglish
Pages (from-to)589-596
Number of pages8
JournalProcedia Computer Science
Volume130
DOIs
StatePublished - 2018

Bibliographical note

Publisher Copyright:
© 2018 The Authors. Published by Elsevier B.V.

Keywords

  • Arabic Language
  • Deep Learning
  • Ontology
  • Semantic Annotation

ASJC Scopus subject areas

  • General Computer Science

Fingerprint

Dive into the research topics of 'Semantic Annotation of Arabic Web Documents using Deep Learning'. Together they form a unique fingerprint.

Cite this