|
|
|
projects
|
|
Zenodotus - Ontological Document Management System
The project is about Ontological Document and Content Management Systems, as described in a publication presented at the 9th International Conference on Business Information Systems (BIS 2006). It currently takes the form of a proof of concept being developed within the university in collaboration with the Faculty of Letters and Human Sciences for a repository of e-learning resources for Modern French.
knOWLer - Ontology-based Information Management System
knOWLer is an ontology-based information management system targeting semantic integration into large-scale information systems. The semantic is provided through an ontology language (OWL), showing that ontological reasoning can be scaled to sizes of standard IR systems.
ICDBrowser - Intelligent Browser for the ICD 10
The goal of the projet is to develop a classification tool for ICD10, using ontological information to dynamically structur results of requests. A prototype to code patients's files with CIM 10 has been developed (in French only).
IKARO - Improved Knowledge Access through Repurposed Ontologies
The project aims to facilitate access to ontologies by restructuring their concepts dynamically according to the actual application. Our approach is to create on top of the original ontology a virtual ontology whose structure is defined explicitly or learned implicitly. There is already a prototype for classifying medical services with TARMED.
SplitTrees - Formalization and Comparison of Split Criteria for Decision Trees
This project is about decision trees in the context of large data sets. To achieve more fundamental insights on how to select split criteria, we propose a formal methodology that permits to compare multiple split criteria. We introduce a new family of split functions, which have a more stable behavior than the classical split criteria when larger and larger data sets are used to construct decision trees. Our new technique of sampling applied to large data sets guarantees that the decision trees inferred from the whole database as well as from the sampled database are almost always identical.
HTS Classifier
In this project we provide web-based access to the concepts and attributes of the HTS Ontology, with the help of a light-weight knowledge-representation system.
|
|
