MACHINE LEARNING AND KNOWLEDGE ACQUISITION
This course describes the techniques of nonsupervised learning developed within the area of machine learning.
In the part on supervised learning, we will deal with the learning of concepts as a search problem, inductive learning, analytic learning and
case-based reasoning.
In the case of nonsupervised learning, an overview of the problem is given from the different areas that have been addressed (cognitive
psychology, numerical taxonomy, data analysis) and techniques that have been developed.
We will then go on to explore the different features that appear in the area of learning, such as conceptual grouping and the formation of
concepts and the different systems that have been developed.
Finally, there is a detailed study of the problems and elements of a nonsupervised learning system using the LINNEO system as an
example of a learning tool, and different alternatives are examined.
- 1. Introduction.
- 2. Learning of concepts and the general/hierarchical hierarchy.
2.1 Learning of concepts as a search problem.
- 2.2. Version space.
- 2.3. Inductive bias.
3. Induction.
- 3.2. Decision trees.
- 3.2. Advanced features of decision trees.
- 3.3 The Icarus cognitive architecture.
4. Analytic learning.
4.1. Explanation-based generalization. - 4.2. The Prodigy architecture.
- 4.3. The Soar architecture.
- 5.
Case-based learning.
- 5.1. Representation of cases.
- 5.2. Indexing and recovery.
- 5.3. Precedent adaptation.
- 5.4. Analogy.
- 5.5. The Chef
system.
- 5.6. The Casey system.
Advisors
- BEJAR ALONSO, JAVIER
- CORTES GARCIA, CLAUDIO ULISES
- SANCHEZ MARRE, MIQUEL
- SANGÜESA I SOLE, RAMON
Bibliography
- Dubes & Jain. Algorithms for Clustering Data. Prentice Hall 1988
- D. Leake (Ed.). Case-Based Reasoning. Experiences, Lessons, Future
Directions. The MIT Press, 1996.
- Fisher, Pazzani, Langley Eds. Concept Formation: Knowledge and
Experience in Unsupervised Learning
- Fisher. Knowledge Acquisition via incremental Conceptual Clustering.
Machine Learning 20:2-22, 1987
- I. Watson. Applying Case-Based Reasoning. Techniques for Enterprise
Systems. Morgan Kaufmann, 1997.
- J. Kolodner. Case-Based Reasoning. Morgan Kaufmann, 1993.
- K.D. Althoff, E. Auriol, R. Barletta & M. Manago. A Review of industrial Case-Based Reasoning tools. An AI Perspectives Report, AI Intelligence, 1995.
- LNAI serie de Agentes Inteligentes 1 a 6
- Michalski & Stepp. Learning from observation: Conceptual Clustering.
- Machine Learning: An AI approach. Michalski, Carbonell, Mitchell
(Eds). Springer 1984
- Pat Langley. Elements of Machine Learning. Morgan Kaufmann. 1996
- Procc. of EWCBR'94, EWCBR'96, EWCBR'98, EWCBR'2000 Workshops
- Procc. of ICCBR'95, ICCBR'97, ICCBR'99 Conferences
- Smith & Medin. Categories and Concepts. Harvard University press1981