Concept-based semantic annotation, indexing and retrieval of office-like document units
Saša Nešić, Fábio Crestani, Mehdi Jazayeri, Dragan Gašević · 2010
The paper presents an ontology-driven approach to semantic annotation, indexing and retrieval of fine-grained units of document data. In this approach, the document units and the user query are both represented by weighted vectors of ontological concepts. To determine the relevance of the document units to given query, we measure the similarity between their concept vectors. Results of the preliminary evaluation conducted with a prototype implementation are promising. We present brief analysis of these results.