Dr. Florence Amardeilh - R&D Director

University Paris X Nanterre – MoDyCo

Thesis: Semantic Web and Computational Linguistic: methodological proposals and conception of a software platform.

This thesis deals with the issues related to semantic annotation and ontology population within the framework defined by the Semantic Web. The idea suggested here is to combine the Information Extraction tools with the SW knowledge representation tools, such as ITM from Mondeca. But there is currently a gap between the representation formats used by these tools. This thesis proposes to fill this gap by designing a mediator able to:

• Semi-automatically create of multilingual domain vocabularies/terminologies from a representative corpus of textual documents

• Semi-automatically enrich a knowledge base with the named entities and semantic relations extracted from the textual documents

• Semantically annotate those documents with the concepts modelled by the reference ontology

Keywords: Semantic Web, Computational Linguistic, Information Extraction, Knowledge Representation and Acquisition, Ontologies, Semantic Annotation.

MoDyCo's home page

Bernard Vatant - Senior Consultant

Expertise : Knowledge Engineering, Semantic Web languages and technologies, Ontology Modelling
Background : Mathematics teacher, graduated from ENSET (1975)

Activities in the standards area

W3C Semantic Web Activity

• WebOnt (OWL) Working Group (2002-2003)
• Semantic Web Best Practices and Deployment Group (2004-2005)
• Collaboration to SKOS (since 2005)

ISO Topic Maps

• Topic Maps XTM working group (2000-2001)
• OASIS Topic Maps Published Subjects Technical Committee (Chair 2001-2003)

Home Page : univers immedia http://perso.orange.fr/universimmedia/index.htm

Blogs : Leçons de Choses http://mondeca.wordpress.com/

The wheel and the hub http://universimmedia.blogspot.com/

Dr Hondjack Dehainsala - R&D Engineer

University of Poitiers – LISI-ENSMA

My thesis was to propose a new database model called Ontology-Based Database (OBDB) which has two main characteristics.

• First it allows to manage both ontologies and data in a same repository
• Second, it permits to associate each data with the ontological concept which defines its meaning.

Our approach addresses ontologies expressed as models, i.e. sets of objects that may be accessed from a root object, and that consists of classes and properties. Target data are those data that are instances of ontology classes and that fulfil a strong typing assumption : each instance shall belong to exactly one base class, defined as the lower bound (for the subsumption relationship) of the classes the instance belongs to, and it shall be described only by properties that are applicable to its base class. We have proposed an OBDB architecture model, called OntoDB and implementation a proposed a prototype using the PostgreSQL DBMS and the PLIB ontology model. The OntoDB architecture has three characteristics :

1. - it offers data access, at the ontology level, whatever is the type of the used DBMS (relational, object-relational or object oriented).
2. - it supports evolutions of the used ontologies (adding new classes, new properties, etc.) and their population schemas, and
3. - it offers data access, at the ontology level, whatever is the type of the used DBMS (relational, object-relational or object oriented).

Keywords: Databases, Ontologies, Ontology-Based Database, PLIB, RDF Schema, OWL.

LISI-ENSMA homepage

Olivier Carloni - PhD Student

Université de Montpellier - LIRMM, France

The purpose of this Ph.D. is to provide reasoning capabilities to ITM, the knowledge base management system designed by Mondeca. One characteristic of this work is to build these reasoning mechanisms using graph theory operations. At the end, the reasoning system should permit:

(1) import and export of knowledge serialized using web semantic standards (more precisely Topics Maps (ISO standard)).

(2) knowledge internal ITM description, storage and management using ITM labeled hypergraph formalism,

(3) the full processing of this knowledge using conceptual graphs formalism.

LIRMM's home page

Laurence Noel - PhD Student

This Ph. D. thesis aims at exploiting semantically annotated data to improve user experience concerning information access and knowledge acquisition by the end-user. This work will attempt to define articulation modes between two types of systems : one based on a formal representation of knowledge and the other one based on a hypermedia representation of information. The goal is to answer the following user needs:

• exploring and discovering the information space (transversal navigation and exploratory search)
• comparing the elements in this information space (information hierarchisation and contextualisation of data according to four display modes: spatial, temporal, thematic, and iconic)
• selecting and reorganizing those elements according to personal criteria (interface adaptation and data appropriation)

Keywords: Semantic Web User Interface, Hypermedia, Interaction design, Information architecture, Information design, User Experience

LEDEN's home page

Pierre-Yves Vandenbussche - PhD Student

University Paris 6 – SPIM INSERM

Thesis: The design and the implementation of an Information resources meta-model and its services : application to the medical field.

Considering a specific field, information resources are available in different organizational systems: ontological, terminological, thesauri... Information resources design in a structured formal method will improve:

• interoperability among resources and software applications, e.g. input/output of this framework (APIs, web services, serialisations...) as an application server or as a referential.

• knowledge segmentations in levels of expression, e.g. depending on the complexity, the formalisation of areferential and their interactions. This segmentation will offer more significant services in regard of the level aimed.

Keywords: Semantic Web, Meta-model, Ontologies, Information resources, Interoperability, Expression levels.

SPIM website