Name
Abstract | ||
|---|---|---|
One approach to deal with large query answers or large collections of text documents is to impose some kind of structure to the collection for instance by a grouping into clusters of somehow related or close items. Another approach is to consider characteristics of the collection for instance by considering central and/or as a frequent keywords possibly taken from a background vocabulary or a more thorough structuring of background knowledge, like taxonomies or ontologies. In this paper we present a preliminary approach to combine these directions. More specifically we address an approach where conceptual summaries can be provided as answers to queries or survey over a document collection. The general idea is to apply a background knowledge ontology in connection with a combined clustering and generalization of keywords. Preliminary experiments with Wordnet as background knowledge and excerpts from Semcor as data are presented and discussed. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1007/978-3-540-68123-6_54 | ISMIS |
Keywords | DocType | Volume |
background knowledge ontology,combined clustering,background vocabulary,hierarchical clustering,preliminary experiment,large query answer,document collection,ontological summary,large collection,preliminary approach,close item | Conference | 4994 |
ISSN | ISBN | Citations |
0302-9743 | 3-540-68122-1 | 0 |
PageRank | References | Authors |
0.34 | 11 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Troels Andreasen | 1 | 505 | 43.70 |
| Henrik Bulskov | 2 | 152 | 15.95 |
| Thomas Vestskov Terney | 3 | 1 | 1.41 |