Robust Selection of Domain-specific Semantic Similarity Measures from Uncertain Expertise - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Robust Selection of Domain-specific Semantic Similarity Measures from Uncertain Expertise

Résumé

Knowledge-based semantic measures are cornerstone to exploit ontologies not only for exact inferences or retrieval processes, but also for data analyses and inexact searches. Abstract theoretical frameworks have recently been proposed in order to study the large diversity of measures available; they demonstrate that groups of measures are particular instantiations of general parameterized functions. In this paper, we study how such frameworks can be used to support the selection/design of measures. Based on (i) a theoretical framework unifying the measures, (ii) a software solution implementing this framework and (iii) a domain-specific benchmark, we define a semi-supervised learning technique to distinguish best measures for a concrete application. Next, considering uncertainty in both experts' judgments and measures' selection process, we extend this proposal for robust selection of semantic measures that best resists to these uncertainties. We illustrate our approach through a real use case in the biomedical domain.
Fichier non déposé

Dates et versions

hal-01059564 , version 1 (01-09-2014)

Identifiants

Citer

Stefan Janaqi, Sébastien Harispe, Sylvie Ranwez, Jacky Montmain. Robust Selection of Domain-specific Semantic Similarity Measures from Uncertain Expertise. IPMU 2014, 15th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, Jul 2014, Montpellier, France. pp.1-10, ⟨10.1007/978-3-319-08852-5_1⟩. ⟨hal-01059564⟩
21 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More