Multi-label classification method and system based on fireworks algorithm

Abstract

The invention discloses a multi-label classification method and a multi-label classification system based on a fireworks algorithm. The method specifically comprises the steps of acquiring a label ofa known training sample, wherein the training sample is provided with a plurality of training sample nodes; performing multi-label classification on a to-be-predicted sample according to the label ofthe training sample, and calculating an optimal characteristic weight value between the to-be-predicted sample and the training sample by use of the fireworks algorithm; calculating a weighted Euclidean distance between the to-be-predicted sample and the training sample according to the optimal characteristic weight value; acquiring K-nearest nodes nearest to the to-be-predicted sample from the plurality of training sample nodes according to the weighted Euclidean distance; and acquiring the label of the to-be-predicted sample according to the label of the k-nearest nodes. The optimal characteristic value in a classification algorithm is calculated by use of the fireworks algorithm, and thus the precision of the multi-label classification is improved.

Description

Technical field [0001] The invention relates to the field of multi-label classification, in particular to a multi-label classification method and system based on a firework algorithm. Background technique [0002] For multi-label classification problems, the current main solutions are divided into problem conversion and algorithm adaptation. The basic idea of ​​problem conversion is to process multi-label training samples to convert multi-label classification problems into known learning problems. . [0003] A multi-label classification algorithm based on particle swarm optimization algorithm is proposed in the prior art, including an optimization stage and a classification stage. The optimization stage uses the particle swarm algorithm to optimize the feature weights of the feature weighted nearest neighbor classification algorithm, and the classification stage optimizes The feature weights obtained in the stage are applied to the feature weighted nearest neighbor classification ...

AI Technical Summary

Problems solved by technology

[0004] But when the samples are unbalanced, for example, when the sample size of one class is large, while the sample size of other classes is small, it may cause that when a new sample is input, the samples of the large-capacity class in the neighbors of the input sample account for the majority, so It will cause the new sample not close to the target sample, or the new sample is very close to the target sample, the particle swarm optimization algorithm tends to converge to the local optimal solution, resulting in low accuracy of multi-label classification

Images