Aristotle University Of Thessaloniki (AUTH)
Μέτρα ομοιότητας με εναλλακτικές μεθόδους για συστήματα συλλογιστικής των περιπτώσεων (Case Based Reasoning)
Abstract
dc:descriptionCase-based Reasoning systems constitute sector of “Lazy” Machine Learning systems . They reach the learning with the storage of older experiences and during the recent years they have known big growth. The use of a similarity function, which is used for the retrieval of most similar case from the Case Base, is a critical point for the operation of Case-based Reasoning systems. Classic resemblance metric that is used in the established Case-based Reasoning systems include the Euclidean distance and its variants. Alternative methods can be applied in connection with the nature of data and the particularities of the problem. In this thesis we present and study non established techniques of resemblance that they can be used in systems based on cases. Such approaches, which include using fuzzy logic and the edit distance function, are described and experiments are executed with systems that use them. Combination of use of more than one techniques in case based reasoning systems is also proposed.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2004
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Νικολαΐδης, Σάββας
Subjects
dc:subject × 14- Συλλογιστική των περιπτώσεων
- Μηχανική μάθηση
- Συστήματα γνώσης
- Τεχνητή νοημοσύνη
- Επιστήμη υπολογιστών
- Case based reasoning
- Machine learning
- Knowledge based systems
- Artificial intelligence
- Computer Science
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/14587