Aggregating Similarity Measures based Ontology on Documents Retrieval

Abstract

This paper investigates a methodology for the ontology based semantic retrieval of annotated web documents with terms occurrence weighting. The semantic structural distance of document terms in terms of domain ontology is computed against new unknown queries to improve the documents ranking and retrieval. Furthermore, the role of aggregation methods to combine the weighting terms scheme based similarity measures with the similarity of semantic distance with respect to ontology have been carried out. Domain ontology is developed in which it defines a number of keywords and properties for semantic reasoning. The performance of the proposed method has been evaluated on health documents from a number of aspects, and experimental results are encouraging showing its effectiveness.