Cases as terms:
A feature term approach to the structured
representation of cases

Enric Plaza*
Institut dInvestigaci en Intel.ligncia Artificial, CSIC
Campus de la Universitat Autnoma de Barcelona
08193 Bellaterra, Catalunya, Spain
Email: enric@iiia.csic.es
URL: http://www.iiia.csic.es

1 Motivation

In our research work, we have come to represent cases as complex, structured data
structures that we will formally described as feature terms (see [Arcos] for a description of
the NOOS language for CBR and multistrategy leaming). The advantage of using
structured representations are twofold. Firstly, it offers a natural way to describe composite
objects that if described by attribute-value representations cause some problems like dealing
with irrelevant attributes, not-applicable values, etc (and this leads to problems when
comparing similitude among cases in this descriptions). Secondly. structured-
representation cases offer the capability of treating subparts of cases also as full-fledged
cases: they can be stored, and retrieved and used to solve (sub)problems of new cases;
also, a case may be solved using (subparts ot) multiple cases retneved from the systems
memory. We will present here a formalization of structured (sometimes called object-centered) 
representations as feature terms and we will present how can we assess similarity
between feature terms (cases) and determine the preferred (most similar) case from a set of
cases.
References
[At-Kaci] H At-Kaci, A Podelski (1992), Towards a meaning of LIFE. PRL Research
Report #11, Digital Research Laboratory (available at doc-server@prl.dec.com sending
a message with subject line help).

[Arcos] Arcos, J. L., and Plaza, E. (1993), A Reflective Architecture for Integrated
Memory-based Learning and Reasoning, In S. Wess, K.D. Althoff, M.M. Richter (Eds.),
Topics in Case-Based Reasoning. Lecture Notes in Artificial Intelligence, 837, p. 289-300.
Springer Verlag: Berlin.

[Armengol] Armengol, E. and Plaza, E. (1994b), Integrating Induction in a Case-based
Reasoner. Proceedings of the Second European Workshop on Case-based Reasoning.
Chantilly (France). pp. 243-251.

[Brner] Bomer, K (1994), Structural similarity as a guidance in case-based design. In S
Wess, K D Althoff, M M Richter (Eds.) Topics in Case-Based Reasoning, Lecture Notes
in Artificial Intelligence. Vol. 837, p.197-208. Springer-Verlag 1994.
[Bunke] Bunke, H and Messmer, B T (1994), Similarity measures for structured
representations. In S Wess, K D Althoff, M M Richter (Eds.) Topics in Case-Based
Reasoning, p. 106-118. Lecture Notes in Artificial Intelligence 837, Springer Verlag.

[Carpenter] Carpenter. B (1992). The Logic of Typed Feature Structures. Cambridge
Tracts in Theoretical Computer Science. Cambridge University Press, Cambridge, UK.
[deRaedt] deRaedt, L (1992), Interactive Theory Revision. Academic Press: London.
[Kettler] B P Kettler, J A Hendler, W A Anderson, M P Evett (1994), Massively parallel
support for case-based planning. IEEE Expert, p. 8-14, Fed. 1994.

[Jantke] K P Jantke (1993), Nonstandard concepts of similarity in case-based reasoning.
Proceedings of the 17th Annual Conference of the Gesellllschaft fr Klassifikation e. V. .
Kaiserslautern, March 1993. Springer Verlag.

[Lpez] B. Lpez, E. Plaza (1991), Case-based Learning of Strategic Knowledge. Lecture
Notes in Artificial Intelligence 482. Springer-Verlag.1991, pp. 398-411.
[Plaza] Plaza, E., Arcos, J.L. (1994), Integration of learning into a knowledge modelling
framework. Lecture Notes in Artificial Intelligence, Vol. 867. Springer-Verlag 1994, pp.
355-373. Available online at URL http://www.iiia.csic.es/People/enric/EKAW-
94_ToC.html

[Plotkin] Gordon D Plotkin, A note on inductive generalization. In B. Meltzer and D
Michie (Eds.), Machine Intelligence 5, p. 153-163. Elsevier 1970.

[Russell] Russell, S. (1990), The Use of Knowledge in Analogy and Induction. Morgan
Kaufmann.
