The usage of ontologies for annotating resources significantly growth in the last years. However, ontologies are dynamics artifacts that may change dur- ing time due to the change of the conceptual representation of the domain they describe or to the necessity of modifying the definition of some entities. The us- age of repositories containing annotated resources available for users, like in the web environment, leads to the necessity of managing the changes carried on the ontology in order to avoid both the decrease of effectiveness of the search sys- tem and the change, by the users, of how they have organized their knowledge. In this work, we have investigated how ontology evolution operations impact on the effectiveness of search systems. We have performed the evaluation by imple- menting an ontology versioning approach that permits to fully track the changes carried out on the ontology and by performing an information retrieval evalua- tion by using two different document collections, the MuchMore collection and the TREC 2004 Genomic Track collection.
Evaluating the Impact of Ontology Evolution Patterns on the Effectiveness of Resources Retrieval
Dragoni, Mauro;Ghidini, Chiara
2012-01-01
Abstract
The usage of ontologies for annotating resources significantly growth in the last years. However, ontologies are dynamics artifacts that may change dur- ing time due to the change of the conceptual representation of the domain they describe or to the necessity of modifying the definition of some entities. The us- age of repositories containing annotated resources available for users, like in the web environment, leads to the necessity of managing the changes carried on the ontology in order to avoid both the decrease of effectiveness of the search sys- tem and the change, by the users, of how they have organized their knowledge. In this work, we have investigated how ontology evolution operations impact on the effectiveness of search systems. We have performed the evaluation by imple- menting an ontology versioning approach that permits to fully track the changes carried out on the ontology and by performing an information retrieval evalua- tion by using two different document collections, the MuchMore collection and the TREC 2004 Genomic Track collection.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.