EUVIMEDEUVIMED entdecken
Crossref

Soundness, Completeness, and Compactness Theorems and Tableaux Calculi for The Madme Ontology

Simone Cuconato

Romanian Journal of Mathematics and Computer Science · 2025

DOI: 10.66899/17b0eb45 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Romanian Journal of Mathematics and Computer Science, 2025. Verfasst von Simone Cuconato. Die Arbeit wurde bisher 1-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper bridges mathematical logic and data science by demonstrating three important theorem – soundness, completeness, and compactness – of MADME (MAke Decision for Metadata Extraction), a logic-based decision-making ontology for structured metadata extraction. We formally establish these properties to ensure MADME’s logical consistency, expressiveness, and computational efficiency. Building on this formal foundation, we integrate tableaux calculi to enhance inference: this extension significantly improves extraction accuracy, deepens the interpretability of decision steps, and thus increases the overall reliability of the extracted metadata. By combining formal logic verification with tableau-driven reasoning, MADME delivers a scalable, transparent, and robust framework for metadata extraction in data science.

Zitieren

APA

Cuconato, S. (2025). Soundness, Completeness, and Compactness Theorems and Tableaux Calculi for The Madme Ontology. Romanian Journal of Mathematics and Computer Science. https://doi.org/10.66899/17b0eb45

Vancouver

Cuconato S. Soundness, Completeness, and Compactness Theorems and Tableaux Calculi for The Madme Ontology. Romanian Journal of Mathematics and Computer Science. 2025. doi:10.66899/17b0eb45

BibTeX
@article{Cuconato2025,
  author = {Cuconato, Simone},
  title = {Soundness, Completeness, and Compactness Theorems and Tableaux Calculi for The Madme Ontology},
  journal = {Romanian Journal of Mathematics and Computer Science},
  year = {2025},
  doi = {10.66899/17b0eb45}
}
Kostenlos starten →