An expert system for speech decoding
Selected and updated papers from the proceedings of the 1982 European conference on Progress in artificial intelligence
Artificial intelligence (2nd ed.)
Artificial intelligence (2nd ed.)
An interference matching technique for inducing abstractions
Communications of the ACM
A Framework for Representing Knowledge
A Framework for Representing Knowledge
Version spaces: an approach to concept learning.
Version spaces: an approach to concept learning.
Subjective bayesian methods for rule-based inference systems
AFIPS '76 Proceedings of the June 7-10, 1976, national computer conference and exposition
A synthetic view of approximate reasoning techniques
IJCAI'83 Proceedings of the Eighth international joint conference on Artificial intelligence - Volume 1
Induction of concepts in the predicate calculus
IJCAI'75 Proceedings of the 4th international joint conference on Artificial intelligence - Volume 1
Learning by creatifying transfer frames
Artificial Intelligence
| Hi-index | 0.00 |
This paper describes the learning module of the Expert System BIMBO, devoted to the task or understanding continuous speech. Learning is performed in presence of uncertainty and of errors in the data and in the general knowledge. Both learning from samples and by interaction with a human expert are allowed. The system performs constructive generalization and uses a rormal definition of equivalence between groups of rules in order to automatically build up a multilevel partitioned network of production rules. Consistency of the learned knowledge base is checked in a semi-automatical way, according to given consistency criteria. Once the system has built up its initial knowledge, incremental learning is performed, by analyzing the outcomes of the system runs or by trying to insert new knowledge the human expert want to add. The learning module or the system is implemented in Franz Lisp on a VAX-11/780.