Use of Similarity of Wiki Pages as an Instrument of Domain Representation for Semantic Retrieval

Abstract

We propose an ontology-based approach to the Web semantic search that uses thesaurus representation of user task. Domain ontologies are considered as a source of semantic markup of the Wiki resources pertinent to retrieval domain. We use practical aspects of ontological approach to organization of Wiki-based information resources. Ontological model of Wiki resource formalizes the knowledge base structure and explicitly represents its main features. Domain ontologies, Wiki resources and task thesauri are generated independently by different applications but are used in general technological chain of useroriented semantic retrieval. Open information environment is considered as an external data base with great volumes of heterogeneous and semi-structured information. Wiki ontologies are considered as the basis for establishing a semantic similarity between domain concepts pertinent to user task. Such Wiki-ontology elements as classes, property values of class instances and relations between them are used as parameters for the quantitative assessment of semantic similarity.