Towards a Principle-Based Approach for Case-Based Reasoning - Intelligence Artificielle Access content directly
Conference Papers Year : 2022

Towards a Principle-Based Approach for Case-Based Reasoning

Abstract

Case-based reasoning (CBR) is an experience-based approach to solving problems; it adapts previously successful cases to new problems following the key assumption: the more similar the cases, the more similar their solutions. Despite its popularity, there are few works on foundations, or properties, that may underlie CBR models. This paper bridges this gap by defining various notions capturing the above assumption, and proposing a set of principles that a CBR system would satisfy. We discuss their properties and show that the principles that are founded on the CBR assumption are incompatible with some axioms underlying non-monotonic reasoning (NMR). This shows that CBR and NMR are different forms of reasoning, and sheds light on the reasons behind their disagreements.
Fichier principal
Vignette du fichier
Towards a Principle-Based Approach for Case-Based Reasoning.pdf (341.58 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03843724 , version 1 (17-11-2022)

Identifiers

Cite

Leila Amgoud, Vivien Beuselinck. Towards a Principle-Based Approach for Case-Based Reasoning. 15th International Conference on Scalable Uncertainty Management (SUM 2022), Oct 2022, Paris, France. pp.37-46, ⟨10.1007/978-3-031-18843-5_3⟩. ⟨hal-03843724⟩
80 View
62 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More