The Semantic Search Engine is a web-based search engine tool for retrieving information stored in ontology-driven knowledge bases. As part of the Digital Shape Workbench (DSW) infrastructure of the  AIM@SHAPE Network of Excellence its primary objective is the implementation of the necessary infrastructure to support search and retrieval of objects in scientific collections, such as, data sets, simulation models and tools necessary for processing shape objects. Modeling the semantics of shape objects constitutes a concrete step in developing an effective search mechanism for 3D resources.

The Semantic Search Engine can currently be used to search in the various ontologies developed in AIM@SHAPE but could also be used as a front-end for accessing information in ontologies developed outside of the network as well.

Features summary

Version 2 represents the latest prototype version of the program. A brief overview of the features of this version is given below.

Knowledge base

The Semantic Search Engine can handle multiple ontology schemas stored in some conventional database management system that is compatible with the Jena2 ontology management framework.

Ontology language

The ontology language that is currently supported by the Semantic Search Engine is the Web Ontology Language.

Imported ontologies

Supporting imported ontologies is fundamental as most well-developed ontologies (including the AIM@SHAPE ontologies) extend and reuse already existing and well known ontologies which are either generic or address a specific area of interest.

Web-based implementation

The Semantic Search Engine has a web-based implementation to be easily accessible and to present a uniform interface. For your browser to be able to display the Semantic Search Engine you need to install the Java Plug-in first. You can download the plug-in from here.

Inference support

The RacerPro inference engine is utilized by the Search Engine for providing reasoning capabilities. The backend of the Semantic Search Engine (called Ontology Server) manages the interaction between the database system storing the ontology specifications, the inference engine and the Search Engine interface. It essentially loads ontologies in the inference engine, as required, and sets up the communication protocol between clients using the Search Engine interface and the inference engine.

Query formulation

The user is able to formulate graphical queries alleviating much of the typical burdens associated with dealing with query languages. The graphical representation is processed by the Search Engine and the query is translated to the appropriate format ( nRQL) and is then submitted to the inference engine.

Text-based search

A text-based interface is also provided which allows the user to formulate queries that are close to natural language. The user is also able to take advantage of relevant terms found in the AIM@SHAPE Glossary in order to improve the accuracy of the queries.