frlearn.neighbours.classifiers.FRONEC

class frlearn.neighbours.classifiers.FRONEC(Q_type: int = 2, R_d_type: int = 1, k: int = 20, owa_weights: frlearn.utils.owa_operators.OWAOperator = <frlearn.utils.owa_operators.additive object>, nn_search: frlearn.neighbours.neighbour_search.NNSearch = <frlearn.neighbours.neighbour_search.KDTree object>)[source]

Implementation of the Fuzzy ROugh NEighbourhood Consensus (FRONEC) multilabel classifier.

Parameters
Q_typeint {1, 2, 3, }, default=2

Quality measure to use for identifying most relevant instances. Q^1 uses lower approximation, Q^2 uses upper approximation, Q^3 is the mean of Q^1 and Q^2.

R_d_typeint {1, 2, }, default=1

Label similarity relation to use. R_d^1 is simple Hamming similarity. R_d^2 is similar, but takes the prior label probabilities into account.

kint, default=20

Number of neighbours to consider for neighbourhood consensus.

owa_weights: OWAOperator, default=additive()

OWA weights to use for calculation of soft maximum and/or minimum.

nn_searchNNSearch, default=KDTree()

Nearest neighbour search algorithm to use.

References

Ree9072f4f166-1

Vluymans S, Cornelis C, Herrera F, Saeys Y (2018). Multi-label classification using a fuzzy rough neighborhood consensus. Information Sciences, vol 433, pp 96–114. doi: 10.1016/j.ins.2017.12.034

__init__(self, Q_type: 'int' = 2, R_d_type: 'int' = 1, k: 'int' = 20, owa_weights: 'OWAOperator' = <frlearn.utils.owa_operators.additive object at 0x7f8b1a153150>, nn_search: 'NNSearch' = <frlearn.neighbours.neighbour_search.KDTree object at 0x7f8b1a153110>)[source]

Initialize self. See help(type(self)) for accurate signature.

class Model(classifier, X, Y)[source]