jcolibri.method.retrieve.selection.diversity
Class BoundedGreedySelection
java.lang.Object
jcolibri.method.retrieve.selection.diversity.BoundedGreedySelection
public class BoundedGreedySelection
- extends java.lang.Object
Tries to reduce the complexity of the greedy selection algorithm first selecting
the best b·k cases according to their similarity to the query and then applies
the greedy selection method to these cases.
See:
B. Smyth and P. McClave. Similarity vs. diversity. In ICCBR '01: Proceedings
of the 4th International Conference on Case-Based Reasoning, pages 347-361,
London, UK, 2001. Springer-Verlag.
- Version:
- 1.0
- Author:
- Juan A. Recio-Garcia, Developed at University College Cork (Ireland) in collaboration with Derek Bridge.
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
BoundedGreedySelection
public BoundedGreedySelection()
boundedGreddySelection
public static java.util.Collection<CBRCase> boundedGreddySelection(java.util.Collection<RetrievalResult> cases,
CBRQuery query,
NNConfig simConfig,
int k,
int bound)
- Executes the algorithm
- Parameters:
cases
- to retrieve fromquery
- to comparesimConfig
- is the knn config that defines the k returned casesbound
- to create the bounded retrieval setk
- is the number of cases to select
- Returns:
- k cases