Multiple kernel learning classification method based on noise probability function

Abstract

The invention discloses a multiple kernel learning classification method based on a noise probability function. The method comprises the following steps: calculation of the noise probability function; selection of a base classifier ft<*>(x) and calculation of corresponding coefficients at<*> in each round of iteration; and update of weight. The multiple kernel learning classification method based on the noise probability function is suitable for a classification algorithm of a noise-polluted data set, and has the advantages of no need to solve the complex optimization problem, being small in calculation amount compared with a conventional multiple kernel learning method, effectively solving the problem of noise sensitivity of conventional multiple kernel boosting learning, and being better in robustness.

Description

technical field [0001] The invention relates to a multi-core learning classification method based on a noise probability function and belongs to the technical field of data mining. Background technique [0002] Linear Support Vector Machine (SVM) was proposed by Cortes and Vapnik. With the deepening of SVM research, SVM has penetrated into many fields of machine learning, such as pattern classification, regression estimation, probability density estimation, etc. SVM has achieved great success, but it belongs to Single Kernel Learning and has certain limitations. [0003] In the field of machine learning, multiple kernel learning (Multiple Kernel Learning) has attracted more and more attention, because compared with single-kernel learning, multi-kernel learning can overcome the huge scale of sample features, heterogeneous information, multi-dimensional data irregularity and data distribution in high-dimensional feature space. uneven phenomenon. [0004] In recent years, a v...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a multi-core learning classification method based on the noise probability function in order to solve the above problems. This method does not need to solve complex optimization problems, has a small amount of calculation, and effectively solves the problem of sensitivity to noise.

Images