| 摘要 |
<p>Embodiments of the present invention relate to computer-implemented methods and system for ranking named entities that are associated with selected items of content, such as text-based documents (D1-D3), which are stored on the World Wide Web. In one form a method according to an embodiment of the invention in general involves a computer-implemented procedure for ranking entities that are associated with selected items of content (D1-D3), the method generally comprising selecting items of content from a content repository (s710), applying weightings to at least some of the entities of the selected items (S715), wherein the values of at least some of those weightings are influenced by a so-called context, and ranking the entities according to their applied weightings (S720). According to embodiments of the invention, a context may include a search query, a user interest profile, an ontology, or a combination of any one or more of the foregoing. The use of context aims to provide a more meaningful set of entities to present to a user.</p> |
