Format : 14,5x20,5 Reliure : Broché Nombre de pages : 112 Année de parution : 2011 Référence : 95502 I.S.B.N. : 978.2.85428.956.5 Langue :
This special issue of the Information-Interaction-Intelligence (I3) Journal presents a collection of papers on the topic of Complex Object Matching and Discovery.
Applications such as digital libraries, mediation systems and web services-based architectures produce masses of data and describe complex objects such as documents, business processes, services and onto¬logies for different knowledge domains. They are therefore heterogeneous and rich in terms of semantics. Giving the richness of these models and the huge number of stored objects, new techniques and tools are required to help user to retrieve the object satisfying his needs and preferences.
This special issue highlights and explores some of the more recent theoretical and practi¬cal advances in complex object matching and discovery. We received 9 submissions from France, China, Tunisia, Brazil from which 4 papers were accepted. The four selected papers deal with matching and of web services, complex image objects, XML documents and ontology align¬ment. The first paper, by Nadia Yacoubi Ayadi, Malika Charrad, Maria-Esther Vidal, Mohamed Ben Ahmed, Soumaya Amdouni is titled A Semantic Fra-mework for Web service Annotation, Matching and Classification in Bio-informatics. The proposed framework uses a semi-automatic approach, a NLP (Natural Language Processing) driven approach to learn lightweight semantic annotations given a corpus of textual descriptions of Web services.
The authors propose an inference engine to match discovery queries and the knowledge service base comprised by services annotations and background knowledge of the domain. The service classification approach is based on a simultaneous (two-way) clustering algorithm. The second paper by Nadia P. Kozievitch, Ricardo da S. Torres, André Santanchè, Daniel C. G. Pedronette, Rodrigo T. Calumby, Edward A. Fox is titled An Infrastructure for Searching and Harvesting Complex Image Ob¬jects. Complex objects are aggregations of different information combined together into a unique logical object, using sources which may be heteroge¬neous such as metadata, links, images and image manipulation software. The paper presents the design and implementation of a Complex Object-based infrastructure comprising three services : (i) Content-Based Image Retrie¬val (CBIR) (ii) the combination of visual with textual retrieval and (iii) the harvesting of Complex Objects, focusing on the developer view of service integration as components. The third paper, entitled A semantic approach for discovering complex alignments is written by Myriam Lamolle, Chan Le Duc, Rim Touhami. It presents a new approach for discovering complex correspondences bet¬ween atomic concepts and complex concepts that takes into account seman¬tic aspects of entities which are imposed by ontology axioms. First, OWL ontologies are transformed into graphs that represent semantic relationships between ontology entities. Next, several classical aligners are combined to generate correspondences between atomic entities. Finally, in order to gene-rate complex correspondences the method detects relevant subgraphs repre-senting alignable formulae by exploiting their structure and a terminological similarity measure. The fourth paper, entitled TwigStack++ : A New Efficient Holistic Twig Join Algorithm is written by Mohammed Amin Tahraoui and Hamamache Kheddouci. This paper deals with the problem of finding all occurrences of a twig pattern (a small tree) in an XML document. The XML query is defined as a tree and the XML documents are viewed as an ordered tree data. The matching algorithms between twig patterns are based on a two step process, the query is decomposed into elementary relationships (parent-child or ancestor-descendant) and solutions against these pieces are found. In the second phase, all of these partial solutions are joined together to generate the final solutions. This paper proposes a new holistic twig join algorithm to improve these two steps. We wish to thank all researchers who have shown interest in this effort and submitted papers and we also acknowledge the valuable efforts of the revie-wers who had to produce their reviews on a very tight schedule.
Finally, we gratefully acknowledge the editor of the Information-Interaction-Intelligence (I3) Journal, Mohand Boughanem, who was very supportive and helpful in his handing of this issue. Daniela Grigori (PRiSM, University of Versailles St-Quentin en Yvelines, France) Ludovic Lietard (IRISA, University of Rennes 1, France) REVIEWERS Fabio Casati (University of Trento, Italy) Daniela Grigori (PRiSM, University of Versailles St-Quentin en Yvelines, France) Mohand-Saïd Hacid (LIRIS, University of Lyon 1, France) Hamamache Kheddouci (LIESP, University of Lyon 1, France) Geetika T. Lakshmanan (IBM T. J. Watson Research Center, Hawthorne, NY) Ludovic Lietard (IRISA, University of Rennes 1, France) Stéphane Lopes (PRiSM, University of Versailles St-Quentin en Yvelines , France) Hamid Reza Motahari Nezhad, (Hewlett Packard Laboratories, Palo Alto, California) Amedeo Napoli (LORIA, Nancy, France) Andreas Nürnberger (University of Magdeburg, Germany) Olivier Perrin (LORIA, Université Nancy 2, France) Karen Pinel-Sauvagnat, (IRIT, University Paul Sabatier, Toulouse, France)
Editorial Daniela Grigori, Ludovic Lietard
A Semantic Framework for Web service Annotation, Matching and Classification in Bioinformatics Nadia Yacoubi Ayadi, Malika Charrad, Maria-Esther Vidal, Mohamed Ben Ahmed, Soumaya Amdouni
An Infrastructure for Searching and Harvesting Complex Image Objects Nadia P. Kozievitch, Ricardo da S. Torres, André Santanchè, Daniel C. G. Pedronette, Rodrigo T. Calumby, Edward A. Fox
A semantic approach for discovering complex alignments Myriam Lamolle, Chan Le Duc, Rim Touhami
TwigStack++ : A New Efficient Holistic Twig Join Algorithm Mohammed Amin Tahraoui and Hamamache Kheddouci