 |
The ANALOG Project |
The aim of the Analog project is to develop framework
for integrating problem solving methods and learning methods. The approach
followed is developing Noos, a reflective
object-centered representation language that supports modelling and
implementing both problem solving methods and learning methods in an
integrated, seamless way. This research is a follow-up of the Massive Memory
Project.
Currently, learning methods integrated in
Noos include:
- Inductive methods (Antiunification, INDIE, DISC)
- Lazy learning and Case-based Reasoning methods (subsumpion-based
retrieval, preference-based retrieval, similitude terms, similitude relevance)
- EBL and Operationalizationof Noos Methods (PLEC)
- Induction of Noos Methods embodying Functional Equality Restrictions (FERMI)
Currently, implemented problem solving methods include
- Non-linear planning plus case-based learning
- CHROMA - Protein purification (integrates induction and CBR)
- SPIN - Marine Sponges identification (integrates two inductive methods and one
case-based method)
- Reasoning with preferences and higher-order preferences in
non-monotonic tasks
- SAXEX: case-based
learning of musical expressivity in saxophone performance
There is a
Progress Report on the current status of the Analog
project, including relevant articles and reports.
-
Articles
-
Case-based Reasoning with Symbolic Similitudes
- On the Importance of Similitude: An Entropy-based Assessment
- Enric Plaza, Ramon López de Mántaras, and Eva Armengol
- [Compressed
PS |
Compressed
PDF]
-
Inference, Learning and Reflection in Noos
-
Inference and reflection in the object-centered representation language Noos
- and
- Noos:
an integrated framework for problem solving and learning
-
- Cooperation
among Case-based Reasoning Agents
Formalizing Case-based Reasoning
- A Logical
Approach to Case-Based Reasoning Using Fuzzy Similarity Relations
-
Other Publications and Reports
See the Analog Project
Progress Report and more generally see the
IIIA publications and reports.
- For more information
- noos@iiia.csic.es
[Plural]
[Cooperative CBR]
[Federated
Learning]
[Features Terms]
[Noos]
[Team Members]
noos@iiia.csic.es
http://www.iiia.csic.es/Projects/analog/analog-project.html
IIIA