Class HeuristicAverageEmbeddednessRerankerGridSearch<U>
java.lang.Object
es.uam.eps.ir.relison.grid.links.recommendation.rerankers.swap.edge.HeuristicAverageEmbeddednessRerankerGridSearch<U>
- Type Parameters:
U- type of the users
- All Implemented Interfaces:
RerankerGridSearch<U>
public class HeuristicAverageEmbeddednessRerankerGridSearch<U> extends java.lang.Object implements RerankerGridSearch<U>
Grid search for a reranker that optimizes the average embeddedness of the graph using
heuristic approximations.
-
Field Summary
-
Constructor Summary
Constructors Constructor Description HeuristicAverageEmbeddednessRerankerGridSearch() -
Method Summary
Modifier and Type Method Description java.util.Map<java.lang.String,GlobalRerankerFunction<U>>grid(Grid grid)Obtains the different reranking approaches given a grid.java.util.Map<java.lang.String,java.util.function.Supplier<GlobalReranker<U,U>>>grid(Grid grid, int cutoff, java.util.function.Supplier<Normalizer<U>> norm, Graph<U> graph, Communities<U> comms)Obtains the different reranking approaches given a grid.
-
Field Details
-
LAMBDA
private static final java.lang.String LAMBDAIdentifier for the parameter that takes the trade-off between relevance and diversity.- See Also:
- Constant Field Values
-
MODE
private final java.lang.String MODEIdentifier for the update mode of the reranker.- See Also:
- Constant Field Values
-
-
Constructor Details
-
HeuristicAverageEmbeddednessRerankerGridSearch
public HeuristicAverageEmbeddednessRerankerGridSearch()
-
-
Method Details
-
grid
public java.util.Map<java.lang.String,java.util.function.Supplier<GlobalReranker<U,U>>> grid(Grid grid, int cutoff, java.util.function.Supplier<Normalizer<U>> norm, Graph<U> graph, Communities<U> comms)Description copied from interface:RerankerGridSearchObtains the different reranking approaches given a grid.- Specified by:
gridin interfaceRerankerGridSearch<U>- Parameters:
grid- the parameter grid of the algorithm.cutoff- the size of the definitive ranking.norm- suppliers for the normalization scheme.graph- the training graph.comms- the community partition for the training graph.- Returns:
- a map containing suppliers for the different configurations of the reranking approach.
-
grid
Description copied from interface:RerankerGridSearchObtains the different reranking approaches given a grid.- Specified by:
gridin interfaceRerankerGridSearch<U>- Parameters:
grid- the parameter grid for the algorithm.- Returns:
- a map containing suppliers for the different configurations of the reranking approach.
-