We are currently devoloping an integrated business application platform named Corinna3 where the long-term goal is to combine Semantic Web technologies with Natural Language Processing (NLP) to increase the efficiency of the enterprise search process. In this paper we consider one of the Corinna use cases, namely the Human Resource (HR) use case, which has already reached a prototype implementation. While NLP is used for automatic categorization of an unstructured resource, ontologies represent these resources in a taxonomical way with arbitrary properties. On top of the resource ontologies we use both the NLP for the information extraction as well as ontology-based similarity in order to get more acceptable similar search results. Our implemented approach of contextualized hybrid similarity search within the Corinna platform derives strongly from the defined business use cases. It combines and extends existing similarity measures under consideration of additional context information.