We present the SPIRE system, a hybrid case-based reasoning (CBR) and information retrieval (IR) system. It relies on a small case-base to seed the IR system to 1) select documents that are relevant to a presented problem case, and then (2) highlight within these retrieved documents passages that contain relevant infor-mation about specific case features. SPIRE aids not only problem-solving but knowledge acquisition by fo-cusing a text extractor-person or program-on areas of text where needed information is likely to be found. Once extracted, this information can be used to create new cases or data-base objects thus closing the loop in the problem-solving-knowledge-acquisition cycle.

1 lntroduction

When confronted with a new problem, we frequently try to relate it to one that we have dealt with previously. If we have knowledge of a similar prior problem, we might try to apply the same or a related solution. Similarly, we might want to evaluate a scenario and hypothesize the likelihood of various outcomes. By analogizing the new problem, or parts of it, to past experiences, we save our-selves from having to solve every problem from scratch. By relying on prior experiences for both of these processes, we are employing “case-based reasoning". Case-based reasoners solve problems or examine and explain possible outcomes to a scenario by relying on prior similar experiences. Case-base reasoning (CBR) systems can be found in such diverse domains as cooking [8], medical diagnostics [11], manufacturing [10], game playing [12, 4], and legal applications [1, 15].