Capturing semantics from search phrases: Incremental user personification and ontology-driven query transformation

Abstract

Reported is the methodology of the semantic transformation of an initial user's search query in the form of key words or key phrases to the resulting query composed of the relevant concepts of the domain ontology. Transformation methodology is based on incremental user profiling. The mapping of a user's keywords to the concepts of the domain ontology is built according to the presented transformation rules. These rules are based on the usage of the rich set of the semantic relationships comprising subsumption, synonymy, instantiation and meronymy provided as the DAML+OIL ontology. ACM research papers domain is chosen for the methodology evaluation. Transformation algorithm is implemented in the research prototype as the combined capability of the query transformation agent and the ontology agent of the intelligent multi-agent information retrieval mediator1.