Lazy induction of descriptions for relational case-based learning

Abstract

Reasoning and learning from cases are based on the concept of similarity often estimated by a distance. This paper presents LID, a learning technique adequate for domains where cases are best represented by relations among entities. LID is able to 1) define a similitude term, a symbolic description of what is shared between a problem and precedent cases; and 2) assess the importance of the relations involved in a similitude term with respect to the purpose of correctly classifying the problem. The paper describes two application domains of relational case-based learning with LID: marine sponges identification and diabetes risk assessment.