| abstract |
The invention discloses a multi-label classification method and system based on the fireworks algorithm, specifically the labels of the known training samples, the training samples have multiple training sample nodes, and multi-label classification is performed on the to-be-predicted samples according to the labels of the training samples , use the fireworks algorithm to calculate the optimal feature weight between the sample to be predicted and the training sample; calculate the weighted Euclidean weight between the sample to be predicted and the training sample node according to the optimal feature weight distance; according to the weighted Euclidean distance, obtain the k nearest neighbor nodes from the plurality of training sample nodes; according to the labels of the k nearest neighbor nodes, obtain the to-be-predicted The label of the sample. The firework algorithm is used to calculate the optimal feature value in the classification algorithm, which improves the accuracy of multi-label classification. |