Abstract | ||
---|---|---|
This paper discusses the linguistic module of an Ontology Natural Language Interaction System that is based on semantic restrictions. The system, called ONLI, takes as input questions in unrestricted natural language, translates them into nRQL, an extension to the RACER ontology query language, then generates answers as returned by the RACER ontology reasoning server. Translation into nRQL is done through a syntactic analysis (with Minipar), followed by the use of semantic restrictions imposed by the roles stored in the ontology to map terms in the question to concepts and roles in the ontology. The system was evaluated on the FungalWeb ontology using the mean reciprocal rank (MRR) measure used in question-answering. With a test set of 36 questions, the systems achieved an MRR of 0.72. |
Year | Venue | Keywords |
---|---|---|
2006 | FOIS | syntactic analysis,question answering,selectional restrictions,semantic web,mean reciprocal rank,natural language,query language |
Field | DocType | Volume |
Ontology (information science),Ontology-based data integration,Ontology alignment,Information retrieval,Process ontology,Computer science,Ontology Inference Layer,OWL-S,Natural language processing,Artificial intelligence,Suggested Upper Merged Ontology,Upper ontology | Conference | 150 |
ISSN | ISBN | Citations |
0922-6389 | 1-58603-685-8 | 5 |
PageRank | References | Authors |
0.67 | 15 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Leila Kosseim | 1 | 313 | 43.58 |
Reda Siblini | 2 | 8 | 2.46 |
christopher j o baker | 3 | 329 | 30.96 |
Sabine Bergler | 4 | 628 | 80.59 |