Improving query precision using semantic expansion
Information Processing and Management: an International Journal
Content-based peer-to-peer network overlay for full-text federated search
Large Scale Semantic Access to Content (Text, Image, Video, and Sound)
| Hi-index | 0.00 |
Previous research has taught us that the typical non-professional information seeker on the World Wide Web submits very short queries resulting in low-precision results. We show that this behaviour is repeated also by intranet users and therefore apply query expansion (QE) techniques to improve their search results. Arguing that casual searchers are likely to be unwilling to engage in the dialogue required for interactive QE, we provide an automatic QE system based on Latent Semantic Indexing (LSI). Having received mixed results, our analysis suggests that automatic QE based on a collection dependent knowledge structure may work for explorative, i.e. broader, queries whilst targeted and more focused queries suffer from query drift.