Validated Retrieval in Case-Based Reasoning

Abstract

We combine simple retrieval with domain-specific validation of retrieved cases to produce a useful practical tool for case-based reasoning. Based on 200 real-world cases, we retrieve between three and six cases over a wide range of new problems. This represents a selectivity ranging from 1.5% to 3%, compared to an average selectivity of only 11% from simple retrieval alone. c flDigital Equipment Corporation 1990. All rights reserved. 1 1 Introduction We have combined simple retrieval (based on the similarity of surface features) with domain-specific validation of retrieved cases to produce a useful practical tool for case-based reasoning. Starting with a case base of 200 real-world cases, we have narrowed our consideration to between three and six cases over a wide range of new problems. This represents a selectivity ranging from 1.5% to 3%, compared to an average selectivity of only 11% from this same case base using retrieval without validation. We are applying the same technol...

Cite

Text

Simoudis and Miller. "Validated Retrieval in Case-Based Reasoning." AAAI Conference on Artificial Intelligence, 1990.

Markdown

[Simoudis and Miller. "Validated Retrieval in Case-Based Reasoning." AAAI Conference on Artificial Intelligence, 1990.](https://mlanthology.org/aaai/1990/simoudis1990aaai-validated/)

BibTeX

@inproceedings{simoudis1990aaai-validated,
  title     = {{Validated Retrieval in Case-Based Reasoning}},
  author    = {Simoudis, Evangelos and Miller, James},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {1990},
  pages     = {310-315},
  url       = {https://mlanthology.org/aaai/1990/simoudis1990aaai-validated/}
}