In this work, a novel framework for designing Web Information Retrieval systems with particular reference to semantic search engines is presented. The key idea is to add the semantic dimension to the classical Term-Document Matrix thus having a three-dimensional dataset. This enhancement allows for defining a lexico-semantic user interface where the query process is performed at the conceptual level thanks to the use of a Semantic Lexicon. WordNet Semantic Lexicon is used here as golden ontology for handling polysemy and synonymy, hence it is useful for disambiguating user queries at the semantic level. A layered multi-agent system is employed for supporting the design process. Particular emphasis is given to formal system knowledge representation, the interface layer managing user-system interaction and the markup layer performing the semantic tagging process.
On Designing Semantic Lexicon-Based Architectures for Web Information Retrieval / Di Lecce, V; Calabrese, M; Soldo, D. - In: INTERNATIONAL JOURNAL ON ADVANCES IN INTERNET TECHNOLOGY. - ISSN 1942-2652. - ELETTRONICO. - 3:1-2(2010), pp. 170-183.
On Designing Semantic Lexicon-Based Architectures for Web Information Retrieval
Di Lecce V;
2010-01-01
Abstract
In this work, a novel framework for designing Web Information Retrieval systems with particular reference to semantic search engines is presented. The key idea is to add the semantic dimension to the classical Term-Document Matrix thus having a three-dimensional dataset. This enhancement allows for defining a lexico-semantic user interface where the query process is performed at the conceptual level thanks to the use of a Semantic Lexicon. WordNet Semantic Lexicon is used here as golden ontology for handling polysemy and synonymy, hence it is useful for disambiguating user queries at the semantic level. A layered multi-agent system is employed for supporting the design process. Particular emphasis is given to formal system knowledge representation, the interface layer managing user-system interaction and the markup layer performing the semantic tagging process.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.