Section: Application Domains
Keywords : Telecommunications, Multimedia, design of Information Systems (design, information systems), evaluation of Information Systems (evaluation, information system, evaluation, information systems), personalization, ontology, consistency verification.
Evolving Hypermedia Information Systems
We currently focus on web-based information systems (Internet, intranet), or parts of such ISs, offering one of the following characteristics:
-
presence or wanted integration of assistance in collaborative information search and personalization (ranking, filtering, addition of links, etc.);
-
a web-based IS containing information about the activities of a group of people, for instance when they are exchanging verbal or textual information.
-
a web-based IS containing information about the activities of a group of people, for example an institute (INRIA), a company, a scientific community, an European network on the Internet or intranet, etc.
-
a web-based IS indexing a wide range of productions (documents, products) from the Web or a company, according to a thematic criteria, eg. search engines (Yahoo, Voila), Internet guides for specific targets (FT Educadoc) or portals (scientific communities).
-
-
implicit user satisfaction (i.e. the interpretation of the user satisfaction according to the designer point of view) or explicit one, as it is the case for example for business sites, e-learning sites, and also for search engines.
The EPIA RNTL project (2007) is an example of such an evolving hypermedia information system. The Intermed project is an example of the living lab approach.
In summary, our fields of interest are the following:
-
semantic specification and checking of an information system,
-
usage analysis of an information system (Internet, intranet),
-
document mining (XML documents, texts, Web pages),
-
re-designing an information system based on usage analysis,
-
re-designing an information system based on web mining (usage, content and structure),
-
ontology construction and evolution,
-
updating an ontology based on web mining (usage, content and structure),
-
adaptive recommender systems for supporting information retrieval, collaborative search on the Internet,
-
and in general personalization features of an Information System or a service, such as user profiling, personalied interfaces.
Ultimately, it should be noted that other fields (Life Science, Health, Transportation, etc.) could be subject to study since they provide an experimental framework for the validation of our research work in KDD, and in the reuse of experiences managing temporal data; this type of approach may be relevant in applications that are not well solved with methods in Automatics (e.g. nutrition of plants under greenhouses, control in robotics).
