Document relation system based on ontologies for the security domain

Finding semantic similarity or semantic relatedness between unstructured text documents is an ongoing research field in the semantic web area. For larger text corpuses often lexical matching - the matching of shared terms - is applied. Related semantic terms and concepts are not considered in this solution. Also documents that use heterogeneous perspectives on a domain could not be set into a relation properly. In this paper, we present our ongoing work on a flexible and expandable system that handles text documents with different points of view, languages and level of detail. The system is presented in the security domain but could be adapted to other domains. The decision making process is transparent and the result is a ranked list.