"It's not searching, but finding, that counts" - Any amount of information can be queried on the Internet. Since this growing flood of information is increasingly difficult to manage, it becomes ever more important to be able to characterize and describe information with the help of metadata. Semantic search concerns itself with linking this information and using available metadata to improve search results. The possibilities and application scenarios in this space are manifold.
The following advantages can be realized via this solution:
This solution allows content to be automatically categorized and keyworded within the JavaClient as well as WebEdit by initially extracting and analysing page contents. Relevant keywords are then applied to the results based on a customer-defined thesaurus, which are then made available to the editor to apply in whole or in part. The editor is able to initiate a workflow to select and assign keywords.
Keywords are saved both independent of language (according to URIs) as well as language-dependent in freely-definable input components. These input components must be generated in page templates in order to be applied subsequently. Since a template represents a page being generated, this "keywording" is currently only applied at the template level.
In the context of a requirements analysis, or requirements workshop, a determination can be made if it is appropriate to utilize either a unique thesaurus manager or an external solution with a public thesaurus. Subsequently, integration with the separately-licensable PoolParty PowerTagging tool can be carried out as either a cloud-based or on-premise solution. Since downstream search is highly dependent on the target system, an across-the-board integration is not advisable, but would be carried out in the context of the particular project, taking into consideration specific project goals. Similarly, the introduction of a thesaurus would depend on its specific application requirements.