Hierarchical support vector machine classifying method based on rejection subspace

Abstract

The invention relates to a hierarchical support vector machine classifying method based on a rejection subspace. The hierarchical support vector machine classifying method based on the rejection subspace is applicable to processing multi-class or unbalance big data classification problems. The hierarchical support vector machine classifying method is capable of realizing hierarchical parallelization processing on big data in virtue of the rejection subspace so as to improve the classification result. The hierarchical support vector machine classifying method comprises the following steps: firstly, acquiring support vector machines low in computation complexity through training; secondly, determining the rejection subspaces of the support vector machine by virtue of a mutual information learning criterion to obtain rejection training sets in original training sets; and thirdly, training high-accuracy support vector machines on the rejection training sets for further judging the rejection training sets; and the training process is repeated for a plurality of times according to actual requirements. The hierarchical support vector machine classifying method has the advantages that the training complexity of the support vector machine of each layer is reduced according to the idea of dividing and ruling, and the optimal rejection subspace is determined by the data through the mutual information; therefore, the hierarchical support vector machine classifying method has the characteristics of low computation complexity, listening to the data and the like; besides, the method can be applied to the fields of big data classification such as medical diagnosis and multi-class object detection.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition and machine learning, and relates to a hierarchical support vector machine classification method. Background technique [0002] Two common problems in unbalanced big data classification fields such as medical diagnosis and intrusion detection are excessive computational complexity and unbalanced number of class samples. The huge amount of data leads to a great increase in the computational complexity of training classifiers. In addition, the problem of unbalanced number of class samples makes it easy for the classifier to distinguish positive samples as negative samples (in the patent of the present invention, positive samples represent small class samples, and negative class samples represent large class samples). In general, the loss cost of judging a positive sample as a negative sample is significantly higher than the loss cost of judging a negative sample as a positive sample, su...

AI Technical Summary

Problems solved by technology

[0012] The negative sample undersampling method can balance the unbalanced data and reduce the size of the original data set, but this method destroys the internal structure of the original data set, so that the classifier cannot make full use of the original data set. valid information

The positive sample oversampling method can also make the unbalanced data become balanced, but this method makes the scale of the original data set expand again, further increasing the computational complexity of training the classifier

Cost-sensitive learning is an effective method to solve the problem of imbalance in the number of class samples. Some scholars have shown that if the cost of misclassification of the class is known, cost-sensitive learning should be used to solve the problem of imbalance in the number of class samples. However, in the actual classification In the problem, the misclassification cost of the category is often unknown

Images