Personalizing a Concept Similarity Measure in the Description Logic DLELH with Preference Profile

Racharak, Teeradaj; Suntisrivaraporn, Boontawee; Tojo, Satoshi

doi:10.4149/cai_2018_3_581

Cited by 8 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, we define an instance of ∼ (T ,E) for ELH-concept descriptions as follows. Note that other functions apart from the average could be applied; for instance, root mean square and multiplication [4]. We now show that sim satisfies the desirable properties of π ∼ T and the role-depth bounded property as follows.…”

Section: A Computing Interpretable Similaritymentioning

confidence: 74%

“…Recall that a concept similarity measure under preference profile (denoted by π ∼ T ) [4] is a family of functions that maps any pairs of concept descriptions to a real number in [0, 1] under an optional preference expression called preference profile (denoted by π). The higher the value is mapped to, the more likely similarity of them may hold.…”

Section: A Computing Interpretable Similaritymentioning

confidence: 99%

“…When a TBox T is clear from the context, we simply write π ∼. In [4], a good π ∼ T for ELH needs to satisfy the following four properties for any concept descriptions C, D, E, F :…”

Section: A Computing Interpretable Similaritymentioning

confidence: 99%

“…In [4], preferences can be expressed by different means such as an agent's preferences and domain-dependent knowledge. In this work, we consider that preferences are sort of contextual knowledge captured by pre-trained word embeddings.…”

Section: A Computing Interpretable Similaritymentioning

confidence: 99%

“…As a traditional approach on artificial intelligence (AI), similarity of concepts in an ontology can be measured based on the formal semantics of an ontology, especially the modeltheoretic interpretation in DLs. For instance, Racharak et al [4] introduced the measure sim π for the description logic ELH based on the well-known homomorphism characterization of subsumption; Racharak and Tojo [5] also introduced two measures for the description logic FL 0 based on the language-based characterization of subsumption; Lehman and Turhan [6] developed a general framework to instantiate a measure for the same logic ELH; and a dissimilarity measure for the description logic ALC was developed by D'Amato et al [7] based on the well-known tableau method.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

On Approximation of Concept Similarity Measure in Description Logic ELH With Pre-Trained Word Embedding

Racharak

2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

Data-driven and knowledge-driven methods are two mainstream techniques in the pursuit of developing artificial intelligence systems. While data-driven methods seek to develop a decision model from observations in the real world, they are difficult to provide an explanation for the results in human terms. On the other hand, knowledge-driven methods that employ symbolic reasoning based on formal semantics of a knowledge-base are thus more interpretable and explainable, while lacking an ability to deal with incomplete modeling of the structured knowledge-bases. This work aims to tackle these issues on ontology similarity by proposing a general framework that combines the strengths of both approaches for measuring semantic similarity of concepts in a description logic (DL) ontology. More specifically, a neuro-symbolic integrated framework is defined to exploit the pre-trained word embeddings with semantic definitions in an ontology to yield an explainable degree of concept similarity. To demonstrate its applicability, we develop a concrete similarity measure sim conforming to the proposed framework and also introduce an efficient algorithm that can extract an explanation for why such a degree is indicated. The correctness is shown by analyzing theoretical properties that it guarantees to preserve and also by performing an empirical evaluation with a medical ontology SNOMED CT and a medical pre-trained embedding BioWordVec. The results show that our proposed method remains both interpretability and explainability while achieving comparable performance, relative to the state-of-the-art approaches in the data and knowledge-driven methods.

show abstract

Section: A Computing Interpretable Similaritymentioning

confidence: 74%

Section: A Computing Interpretable Similaritymentioning

confidence: 99%

“…When a TBox T is clear from the context, we simply write π ∼. In [4], a good π ∼ T for ELH needs to satisfy the following four properties for any concept descriptions C, D, E, F :…”

Section: A Computing Interpretable Similaritymentioning

confidence: 99%

Section: A Computing Interpretable Similaritymentioning

confidence: 99%