The explanatory power of symbolic similarity in case-based reasoning

A desired capability of automatic problem solvers is that they can explain the results. Such explanations should justify that the solution proposed by the problem solver arises from the known domain knowledge. In this paper we discuss how explanations can be used in case-based reasoning (CBR) in ord...

Descripción completa

Detalles Bibliográficos
Autores: Plaza, Enric, Armengol, Eva, Ontañón, Santiago
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2005
País:España
Institución:Consejo Superior de Investigaciones Científicas (CSIC)
Repositorio:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/162909
Acceso en línea:http://hdl.handle.net/10261/162909
Access Level:acceso abierto
Palabra clave:Case-based reasoning
Lazy learning
Symbolic similarity
Explanations
Descripción
Sumario:A desired capability of automatic problem solvers is that they can explain the results. Such explanations should justify that the solution proposed by the problem solver arises from the known domain knowledge. In this paper we discuss how explanations can be used in case-based reasoning (CBR) in order to justify the results in classification tasks and also for solving new problems. We particularly focus on explanations derived from building a symbolic description of the similar aspects among cases. Moreover, we show how symbolic descriptions of similarity can be exploited in the different processes of CBR, namely retrieve, reuse, revise, and retain. © Springer 2005.