Combining Information Retrieval and Case-Based Reasoning for Middle Ground Text Retrieval Problems

Mike Brown, Christiane Förtsch, and Dieter Wissmann

This paper considers the type of problem for which the potential for amalgamating Information Retrieval (IR) and Case-Based Reasoning (CBR) technologies is the highest. IR is characterised as a bottom-up approach to retrieval of text within unconstrained domains. CBR is characterised as a top-down approach to retrieval of formalised information within domain-specific applications. It is argued that applications that require relatively detailed responses to specific queries of a large-scale, but domain specific text-based archive represent the middle ground between these two disjoint technologies - this is illustrated with a worked example.