Refining Conversational Case Libraries

David W. Aha and Leonard A. Breslow

Navy Center for Applied Research in Artificial Intelligence
Naval Research Laboratory, Washington, DC 20375 USA
{aha,breslow}@aic.nrl.navy.mil



Abstract. Conversational case-based reasoning (CBR) shells (e.g., Inferences 
CBR Express) are commercially successful tools for supporting
the development of help desk and related applications. In contrast to
rule-based expert systems, they capture knowledge as cases rather than
more problematic rules, and they can be incrementally extended. However, 
rather than eliminate the knowledge engineering bottleneck, they
refocus it on case engineering, the task of carefully authoring cases according 
to library design guidelines to ensure good performance. Designing 
complex libraries according to these guidelines is difficult; software is
needed to assist users with case authoring. We describe an approach for
revising case libraries according to design guidelines, its implementation
in CLIRE, and empirical results showing that, under some conditions, this
approach can improve conversational CBR performance.
References

Auriol, E., Wess. S., Manago, M., Althoff, K. -D., & Traphner, R. (1995). INRECA: A
seamlessly integrated system based on inductive inference and case-based reasoning. 
Proceedings of the First ICCBR (pp. 371380). Sesimbra, Portugal: Springer.
Breslow, L., & Aha, D. W. (1997). Simplifying decision trees: A survey. To appear in
Knowledge Engineering Review.
Cardie, C. (1993). Using decision trees to improve case-based learning. Proceedings of
the Tenth ICML (pp. 2532). Amherst, MA: Morgan Kaufmann.
Domingos, P. (1997). Context-sensitive feature selection for lazy learners. Artificial
Intelligence Review, 11, 227253.
Fox, S., & Leake, D. L. (1995). Using introspective reasoning to refine indexing. Proceedings 
of the Fourteenth IJCAI (pp. 391397). Montreal: Morgan Kaufmann.
Heider, R., Auriol, E., Tartarin, E., & Manago, M. (1997). Improving the quality
of case bases for building better decision support systems. In R. Bergmann &
W. Wilke (Eds.) Fifth German Workshop on CBR: Foundations, Systems, and
Applications (Technical Report LSA-97-01E). University of Kaiserslautern, Department 
of Computer Science.
Inference Corporation (1995). CBR2: Designing CBR Express Case Bases. Unpublished.
Kitano H., Shimazu H., & Shibata A. (1993). Case-method: A methodology for building 
large-scale case-based systems. Proceedings of the Eleventh NCAI (pp. 303308). Washington, DC: AAAI Press.
Ku, S., & Suh, Y.-H. (1996). An investigation of the K-tree search algorithm for
efficient case representation and retrieval. Expert Systems with Applications, 11,
571581.
Manago, M., Althoff, K.-D., Auriol, E., Traphoner, R., Wess, S., Conruyt, N., &
Maurer, F. (1993). Induction and reasoning from cases. Proceedings of the First
EWCBR (pp. 313318). Kaiserslautern, Germany: Springer-Verlag.
Nguyen, T., Czerwinsksi, M., & Lee, D. (1993). COMPAQ QuickSource: Providing
the consumer with the power of artificial intelligence. Proceedings of the Fifth IAAI
(pp. 142150). Washington, DC: AAAI Press.
Quinlan, J. R. (1986). Induction of decision trees. Machine Learning, 1, 81106.
Quinlan, J. R. (1989). Unknown attribute values in induction. Proceedings of the Sixth
IWML (pp. 164168). Ithaca, NY: Morgan Kaufmann.
Smyth, B., & Cunningham, P. (1994). A comparison of incremental CBR and inductive 
learning. In M. Keane, 3. P. Haton, & M. Manago (Eds.) Working Papers of
the Second EWCBR. Chantilly, France: Unpublished.
Tan, M., & Schlimmer, J. C. (1990). Two case studies in cost-sensitive concept acquisition. 
Proceedings of the Eighth NCAI (pp. 854860). Boston, MA: AAAI Press.
