Structured induction in expert systems
Structured induction in expert systems
An incremental concept formation approach for learning from databases
Theoretical Computer Science - Special issue on formal methods in databases and software engineering
A Machine-Oriented Logic Based on the Resolution Principle
Journal of the ACM (JACM)
Constructing X-of-N Attributes for Decision Tree Learning
Machine Learning
Machine Learning
Formal Concept Analysis: Mathematical Foundations
Formal Concept Analysis: Mathematical Foundations
Conceptual Knowledge Discovery and Data Analysis
ICCS '00 Proceedings of the Linguistic on Conceptual Structures: Logical Linguistic, and Computational Issues
Duce, an oracle-based approach to constructive induction
IJCAI'87 Proceedings of the 10th international joint conference on Artificial intelligence - Volume 1
Clustering sets of objects using concepts-objects bipartite graphs
SUM'12 Proceedings of the 6th international conference on Scalable Uncertainty Management
AOW '07 Proceedings of the Third Australasian Workshop on Advances in Ontologies - Volume 85
| Hi-index | 0.00 |
We present a method for identifying potentially interesting and useful concepts in a concept lattice and revising the underlying formal context and the lattice it generates to invent new descriptors and extract their definitions. This allows the re-use of concepts in an incremental way. The approach is developed using formal concept analysis and inverse resolution operators for both a theory and its lattice. A consequence of using the concept lattice to represent the concept space is that both unsupervised and supervised approaches are enabled by using different concept evaluation measures. Results are given from experiments in two standard domains with a system called Conduce which implements the method.