Title :
Measuring information content for an ontological concept
Author :
Cross, Valerie ; Chennai-Thiagarajan, Anurekha
Author_Institution :
Comput. Sci. & Software Eng., Miami Univ., Oxford, OH, USA
Abstract :
Numerous ontological or semantic similarity measures have been proposed to determine how similar one concept is to another within the context of an ontology. One category of such measures uses a measure of information content (IC) for ontological concepts. The various approaches to determining IC measures are reviewed. The early corpus-based IC measure is first presented and its weakness discussed. Then the earliest ontology-based IC measure using decedents only is described followed by the two more recent proposals that attempt to improve on the original descendents only ontology-based IC measure. These three IC measures are then analyzed and critiqued based on the assumptions made during their construction. Analysis of the strengths and weakness of the three proposed IC measures suggest intuitive principles to be followed in construction of other IC measures.
Keywords :
information management; ontologies (artificial intelligence); corpus-based IC measure; information content measurement; ontological concept; ontology-based IC measure; semantic similarity measures; Benchmark testing; Biomedical measurements; Correlation; Integrated circuits; Ontologies; Semantics; Standards; corpus-based information content; ontological similarity; ontology-based information content; semantic similarity;
Conference_Titel :
Fuzzy Information Processing Society (NAFIPS), 2012 Annual Meeting of the North American
Conference_Location :
Berkeley, CA
Print_ISBN :
978-1-4673-2336-9
Electronic_ISBN :
pending
DOI :
10.1109/NAFIPS.2012.6291016