mlpy.knowledgerep.cbr.similarity.KMeansSimilarity

class mlpy.knowledgerep.cbr.similarity.KMeansSimilarity(n_cluster=None)[source]

Bases: mlpy.knowledgerep.cbr.similarity.ISimilarity

The KMeans similarity model.

The KMeans similarity model determines similarity between the data in the indexing structure and the query data by using the sklearn.cluster.KMeans algorithm.

Parameters:

n_cluster : int

The number of clusters to fit the raw data in.

Methods

build_indexing_structure(data, id_map) Build the indexing structure.
compute_similarity(data_point) Computes the similarity.