Name
Abstract | ||
|---|---|---|
Multiple convergence is proposed as a method for acquiring disjunctive concept descriptions. Disjunctive descriptions are necessary when the concept representation language is insufficiently expressive to satisfy the completeness and consistency requirements of inductive learning with a single conjunction of generalized features. Multiple convergence overcomes this insufficiency by allowing the disjuncts of a complex concept to be acquired independently. By summarizing correlations among features in the training data, disjunctive concepts can provide rich extensions to the representation language which may enhance subsequent learning. This paper presents the benefits of disjunctive concept descriptions and advocates multiple convergence as an approach to their acquisition. Multiple convergence has been implemented in the learning system HYDRA, and a detailed example of its execution is presented. |
Year | Venue | Keywords |
|---|---|---|
1987 | IJCAI | concept representation language,advocates multiple convergence,concept acquisition,inductive learning,disjunctive concept description,complex concept,multiple convergence,disjunctive description,representation language,subsequent learning,disjunctive concept,satisfiability |
Field | DocType | Citations |
Convergence (routing),Training set,Computer science,Theoretical computer science,Artificial intelligence,Representation language,Completeness (statistics),Machine learning | Conference | 7 |
PageRank | References | Authors |
0.77 | 2 | 1 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kenneth S. Murray | 1 | 78 | 13.86 |