Semantic Relatedness Measure Using Object Properties in an Ontology

Abstract

Abstract. This paper presents a new semantic relatedness measure on ontologies which considers especially the object properties between the concepts. Our approach relies on two hypotheses. Firstly, using only concept hierarchy and object properties, only a few paths can be considered as "semantically corrects" and these paths obey to a given set of rules. Secondly, following a given edge in a path has a cost (represented as a weight), which depends on its type (is-a, part-o f , etc.), its context in the ontology and its position in this path. We propose an evaluation of our measure on the lexical base WordNet using part-o f relation with two different benchmarks. We show that, in this context, our measure outperforms the classical semantic measures.

Effect of Heuristics on Serendipity in Path-Based Storytelling with Linked Data

“…Mazuel and Sabouret [17] suggest to take into account the object property ontology relation between two adjacent items in a path. The base distance measure there is the Jiang and Conrath distance [15], which we can interpret in terms of RDF by looking at the classes of each of the nodes and determining the most common denominator of those classes in the ontology.…”

Semantic Relatedness Measures in E-learning: A study

“…In [2] the authors of the adaptive system My online teacher have developed a method for computing similarity between the concepts by calculating correspondence weight computation between concept attributes. In [8] the authors have proposed a path based semantic relatedness measure using object properties in an ontology. The proposed measures combines the information theoretic approach to semantic similarity, the heterogeneity of relations and the classification of semantically correct and incorrect paths while calculating the semantic relatedness between the concepts.…”

Program understanding using ontologies and dynamic analysis

“…However, the use of multi-relationship paths between ontology concepts to measure semantic similarity is explored in [Mazuel and Sabouret, 2008], an article not cited by the review. In [Mazuel and Sabouret, 2008], paths composed of multiple relations are segmented into single-relation paths. The authors remarked that not all the possible paths between two concepts are semantically correct.…”

“…Accordingly, these paths should not be considered to be connecting the concepts. In Figure 5.6, an example taken from [Mazuel and Sabouret, 2008], the red path should not be used to connect [Mars] and [Water].…”

“…In [Mazuel and Sabouret, 2008], isA and partOf are determined to be an upward and a downward link respectively. The red path in Figure 5.6 is thus semantically incorrect because its pattern UDU is not in the set of semantically correct patterns:…”