Catalogue of Artificial Intelligence Techniques

   

Jump to: Top | Entry | References | Comments

View Maths as: Images | MathML

Case-Based Reasoning

Aliases: CBR

Categories: Inference and Reasoning , Problem Solving , Learning


Author(s): Nicolas Nicolov

Case-based reasoning (CBR) is a problem solving paradigm which utilises the specific knowledge of previously experienced, concrete problem situations (cases). A new problem is solved by finding a similar past case, and reusing its solution in the new problem situation. CBR is a cyclic and integrated process of solving a problem and learning from this experience--central tasks of a CBR system are:

  1. identify the current problem situation (case),
  2. retrieve the most similar stored case (or cases),
  3. reuse the information and knowledge in the retrieved case(s) to solve the new problem--this often involves adapting the old solution to fit the new situation,
  4. revise and evaluate the proposed solution, and
  5. retain the parts of this experience likely to be useful for future problem solving.
Case-based reasoning is claimed to be psychologically plausible, since humans appear to rely heavily on the use of past cases; according to some, expertise is like a library of past experience. CBR is also an approach to incremental, sustained learning, since a new experience is retained each time a problem has been solved, making it immediately available for future problems. When an attempt to solve the current problem fails, the reason for the failure is identified and remembered in order to avoid the same mistake in the future. Case-based reasoning can be considered as a form of Analogical Problem Solving where only information from within the domain is used. CBR systems are restricted to variations on known situations and produce approximate answers but in large domains they will be faster than Rule-based Systems, producing solutions grounded in actual experience. See also Caching.


References:


Comments:

Add Comment

No comments.