Query String Refinement

Created on March 14, 2013, 12:49 p.m. by Hevok & updated by Hevok on May 2, 2013, 5:14 p.m.

Query String Refinement enables more precise Search Results. While the Extension of the Query gets a more complete Result, one can also use the Extensions to get more precise Results where one simply connects the Queries with a logical AND Operator which means all most occur within the Document one is looking for, then one would get much precise Results. The Document Terms are ambiguous and the Problem there is that in the Keyword based Search there are Search Results within the Results that are not really relevant for the Query, because one had something else intended and the Search Terms are ambiguous and one tries to ambiguite the Search Terms to refine the Query String simply by looking up in a Dictionary. There one can look for example for Hypernyms which are generalizations of the Terms one is using. Therefore one can disambiguate Terms with Hypernyms. One can also look up in the Domain Ontology for example Holonyms which are in the part of Relations the Thing where the Query String would be part of a much more general and larger Thing. Then one connect simple for the different Meanings so it would be much more unlikely to find things that refer to something wrong. So one uses these Terms which are mostly Generalizations, Hypernoyms or Holonyms, one connects them with a logical AND with the original Query Search String, then one will get more precise Results. This is Query String Refinement on top of a Traditional Keyword-based Information Retrieval System. The Semantic is only in the Query Extension or Refinement the rest looks like an ordinary, traditional Information Retrieval System.

  • Keyword-based Search does not deliver all Search Results that are relevant for a Query, because Synonyms and Metaphors might describe the queried Content.
  • Extension of the original Query String (Query Extension)

increase precision

query-refinement.png

Tags: precision, improvment, search, semantics
Categories: Concept
Parent: Semantic Search

Update entry (Admin) | See changes

Comment on This Data Unit