ML-kNN cannot be used in real-time classification because of the huge computational cost, time and space resources. Therefore, this paper proposes a LDA-ML-kNN multi-label classification model based on feature dimensionality reduction. Firstly, we use LDA to extract the features of text data, and reduce the dimension of the extracted data matrix, then use data to train the classifiers and add the correlation between the labels. Finally, get the multi-label classification results.
In this paper, the experiment results show that the improved model has a significant improvement in the reduction of computational complexity. When the model has a certain improvement in the average accuracy rate, the compression time and space complexity are greatly reduced. And this model has great practical significance for real-time classification and processing of massive data.