Combining statistical and reinforcement learning in rule-based classification (Q1861616)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Combining statistical and reinforcement learning in rule-based classification |
scientific article; zbMATH DE number 1878617
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Combining statistical and reinforcement learning in rule-based classification |
scientific article; zbMATH DE number 1878617 |
Statements
Combining statistical and reinforcement learning in rule-based classification (English)
0 references
9 March 2003
0 references
The BYPASS (Bayesian Predictive Adaptive Sequential System) algorithm for classification in data mining is considered. BYPASS constructs a collection of classification rules using genetic algorithm techniques: only the rules survive that are frequently ``helpful to the system''. The resulting classification rule is a mixture of all rules in the collection. Results of DNA and multiplexer data analysis are considered.
0 references
Bayesian predictive adaptive sequential system
0 references
clustering
0 references
genetic algorithm
0 references
data mining
0 references
