A Multi-Agents System for Extraction and Annotation of Learning Objects
Aziz Oriche1, Abderrahman Chekry2, Md. Khaldi3

1Aziz Oriche, Abdelmalek Essaâdi Laboratory, LIROSA & LNTE, BP, University, Tetouan Morocco.
2Abderrahman Chekry, Abdelmalek Essaâdi Laboratory, LIROSA & LNTE, BP, University, Tetouan Morocco.
3Mohamed Khaldi, Professor, NUS Normal Upper School Martil, Tetouan, BP, University, Morocco.
Manuscript received on 10 May 2014 | Revised Manuscript received on 20 May 2014 | Manuscript Published on 30 May 2014 | PP: 66-72 | Volume-3 Issue-12, May 2014 | Retrieval Number: L16570531214/14©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: In this paper, we propose an extraction approach of learning objects (LO) (Definition, Example, Exercise, etc…) of documents defined by HTML / XML whose structure is a tree DOM (Document Object Model). Our approach is based on two intelligent agents. An agent of extraction to extract the learning objects independently of the domain and to align them with the concepts of the ontology. An agent of annotation defined by a set of declarative rules for annotates the nodes and their relationships. We defined in the extraction agent a module includes a set of declarative rules of contextual exploration to extract the learning objects contained in the nodes of DOM documents. The result of this extraction is a set of RDF triples generated for alignment the learning objects extracts with concepts of ontology. The agent of annotation is based on a set of annotations metadata representing the results of extraction and annotation. They allow also annotating the neighbor relationship between the nodes. We assume to have a domain ontology defined by concepts, relations between these concepts and properties.
Keywords: Semantic Annotation, Multi-Agent Systems, Ontology, Learning Object (LO), Contextual Exploration Method.

Scope of the Article: Machine Learning