Anatoly Gladun

Semantics-driven modelling of user preferences for information retrieval in the biomedical domain

A large amount of biomedical and genomic data are currently available on the Internet. However, data are distributed into heterogeneous biological information sources, with little or even no organization. Semantic technologies provide a consistent and reliable basis with which to confront the challenges involved in the organization, manipulation and visualization of data and knowledge. One of the knowledge representation techniques used in semantic processing is the ontology, which is commonly defined as a formal and explicit specification of a shared conceptualization of a domain of interest. The work presented here introduces a set of interoperable algorithms that can use domain and ontological information to improve information-retrieval processes. This work presents an ontology-based information-retrieval system for the biomedical domain. This system, with which some experiments have been carried out that are described in this paper, is based on the use of domain ontologies for the creation and normalization of lightweight ontologies that represent user preferences in a determined domain in order to improve information-retrieval processes.

Distant control of student skills by formal model of domain knowledge

We propose to use the domain ontology for student skills examination. Students have to build the personal domain ontologies and thesauri of the examined discipline (based on natural language texts relevant to subject domain – lectures, textbooks, manuals, etc.). These thesauri are then compared with the reference ones built by the tutor. The analysis of student mistakes allows the proposal of personalised recommendations and to improve the course materials in general. Semantic Web technologies are used for knowledge representation and processing for the purpose of interoperability.