Ikram Amous

Semantic rules classification for images annotation

In this paper, we present an approach to facilitate the annotation and the retrieval of the image documents. Our approach is based on the definition and the generation of semantic rules presented via the logic of predicates. Then, we proposed the classification of these rules by a method of clustering Fuzzy C-means. For this, we used the method of co-citation to calculate the similarity measure between images. This classification has grouped thematically these images with the aim to facilitate research and annotation. To validate our proposal, we implemented a tool tested on a set of images of the city center. Finally, we conducted a series of tests to evaluate our approach.

Advanced images research based on an ontological extension

The present paper introduces an approach for image process research. It discusses work in progress and reports the current state of our approach, which comprises the extension of ontology by the semantic rules to facilitate and optimize research. We describe a method for automatic enrichment query SPARQL language.

Modeling the semantic content of the socio-tagged images based on the extended conceptual graphs formalism

With the emergence of Web 2.0, the volume of multimedia documents, particularly the socio-tagged images, has become very considerable. This has made it the annotation and interrogation process of this type of documents a consuming and elusive task. A promising solution is to design methods allowing to model the semantic content of socio-tagged images in order to facilitate their annotation and the research process. Thus, we present in this paper a conceptual modeling approach of the semantic content of these images. First, our idea consists in expanding the conceptual graphs formalism in order to represent the relationships between the concepts and those that are between the concepts and their properties. Second, we use the extended graph to model the semantic content of the socio-tagged images. Experimental studies are conducted on a collection of 25.000 socio-tagged images shared in Flickr. The results demonstrate the effectiveness of our proposed approach.

Adaptation visuelle de documents légataires

Avec l’apparition d’internet, la masse enorme de documents semi-structures induit souvent une surcharge cognitive chez l’utilisateur. Meme si l’utilisateur consulte les memes structures de documents frequemment, il passe son temps a reduire l’affichage pour garder les structures qui lui semblent interessantes et repondant a ses besoins. Pour remedier a ce probleme de surcharge cognitive, differents processus ont ete proposes dans la litterature en adaptant des documents appartenant a des domaines particuliers et construits a priori pour l’adaptation. Compte tenu de notre contexte qui porte sur des documents legataires de type XML non construits a priori pour l’adaptation, nous proposons dans cet article un mecanisme d’adaptation visuelle de ces documents selon les preferences de l’utilisateur.