Kuphi – an Investigation Tool for Searching for and via Semantic Relations

Abstract

In this work, we present a new process-oriented approach for information retrieval called Kuphi. It is intended for investigating entities and their semantic relations to other entities in text documents. We extend the traditional search capabilities which are based on the bag-of-words model in the following way: Starting with a keyword search for a specific entity, the user can not only search for appearances of this entity in the text documents; she can also search via user-specified relations of the queried entity to other entities for these associated entitites in the text. The user has the possibility to search indirectly for manifestations of these relations. Due to cross-lingual annotation, we allow the query language to be different from the language of the documents. We demonstrate our approach with DBpedia as knowledge base and news texts gathered from RSS feeds.