Deriving concept hierarchies from text
Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval
An Information-Theoretic Definition of Similarity
ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Ontology, Metadata, and Semiotics
ICCS '00 Proceedings of the Linguistic on Conceptual Structures: Logical Linguistic, and Computational Issues
Verbs semantics and lexical selection
ACL '94 Proceedings of the 32nd annual meeting on Association for Computational Linguistics
On the Influence of Description Logics Ontologies on Conceptual Similarity
EKAW '08 Proceedings of the 16th international conference on Knowledge Engineering: Practice and Patterns
A generic framework for comparing semantic similarities on a subsumption hierarchy
Proceedings of the 2008 conference on ECAI 2008: 18th European Conference on Artificial Intelligence
Using information content to evaluate semantic similarity in a taxonomy
IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 1
Ontology with likeliness and typicality of objects in concepts
ER'06 Proceedings of the 25th international conference on Conceptual Modeling
Hi-index | 0.00 |
This paper introduces a new similarity measure called SemioSem . The first originality of this measure, which is defined in the context of a semiotic-based approach, is to consider the three dimensions of the conceptualization underlying a domain ontology: the intension (i.e. the properties used to define the concepts), the extension (i.e. the instances of the concepts) and the expression (i.e. the terms used to denote both the concepts and the instances). Thus, SemioSem aims at aggregating and improving existing extensional-based and intensional-based measures, with an original expressional one. The second originality of this measure is to be context-sensitive, and in particular user-sensitive. Indeed, SemioSem is based on multiple informations sources: (1) a textual corpus, validated by the end-user, which must reflect the domain underlying the ontology which is considered, (2) a set of instances known by the end-user, (3) an ontology enriched with the perception of the end-user on how each property associated to a concept c is important for defining c and (4) the emotional state of the end-user. The importance of each source can be modulated according to the context of use and SemioSem remains valid even if one of the source is missing. This makes our measure more flexible, more robust and more close to the end-user's judgment than the other similarity measures which are usually only based on one aspect of a conceptualization and never take the end-user's perceptions and purposes into account.