mlpy.knowledgerep.cbr.features.Feature.retrieval_method

Feature.retrieval_method

The similarity model used during retrieval.

Valid models are:

knn
A k-nearest-neighbor algorithm is used to determine similarity between cases (NeighborSimilarity). The value k must be specified.
radius-n
Similarity between cases is determined by the nearest neighbors within a radius (NeighborSimilarity). The radius n must be specified.
kmeans
Similarity is determined by a kmeans clustering algorithm (KMeansSimilarity).
exact-match
Only exact matches are considered similar (ExactMatchSimilarity).
cosine
A cosine similarity measure is used to determine similarity between cases (CosineSimilarity).
Returns:

str :

The retrieval method.