This thesis makes several contributions to the study of Case-based Reasoning. It presents * a comprehensive description of the foundations of the subject in Knowledge Representation, Machine Learning and Cognitive Science, * A theory of learning for Case-based Reasoning * and it provides a demonstration of this theory by presenting an extensive implementation of a case-based system for Information Retrieval. In the first part of this thesis, research is presented on the foundations of Case-based Reasoning. It relates recent work in Artificial Intelligence to earlier and still developing ideas on the nature of concepts and categories. This part also presents research into the nature of learning in Case-based Reasoning. A review of Schank's Theory of Dynamic Memory is presented and a new Theory of the Acquisition of Episodic Memory is developed. The second part of the thesis is concerned with the practical application of Case-based Reasoning. This research demonstrates how the cognitive processes involved in concept formation and the new Theory of Acquisition of Episodic Memory can be put to practical use. A complete information retrieval system is presented. This system, in addition to being an implementation of the ideas presented in the first part of the thesis, is also intended as a substantive advance in the field of Information Science. It shows how Case-based Reasoning can be used to improve query formulation by exploiting information about the contexts in which queries arise. Particular attention is paid to the problem of recognition of similarity, which is an issue of concern to both Case-based Reasoning and Information Retrieval.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:297351 |
| Date | January 1998 |
| Creators | Guitierrez, Carlos Ramierez |
| Publisher | University of Kent |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | https://kar.kent.ac.uk/21581/ |
Page generated in 0.0035 seconds