Sample label missing data classifier training method

Abstract

The invention discloses a sample label missing data classifier training method which is suitable for processing classified data of two groups of samples, wherein all label data of one group of the samples are lost. An optimized solution technology is provided, label reliability of unlabeled samples is taken as a decision variable to be solved, an optimizing model is established on the basis of a structural risk minimization principle, the model can directly call a nonlinear-programmed toolkit on a small and medium-sized data set for solution, two convex-programmed sub-problems can be solved respectively by an alternate search algorithm on a large-scale data set, and two part variables of the model are solved by means of iteration. The method is high in universality in the different data sets, and good promotion performance on individual test sets is achieved.

Description

technical field [0001] The invention relates to a data analysis method, in particular to a support vector machine-based classifier training method for sample label missing data. Background technique [0002] For the classification data with known two types of sample labels, a more successful data classification method is the support vector machine. This type of method is guided by the principle of structural risk minimization. In a relatively simple function set, the minimum empirical Loss. For nonlinear classification problems, kernel functions are usually introduced to overcome the curse of dimensionality to a considerable extent. Support vector machines have become an important class of data processing and analysis methods for supervised learning problems. However, many practical applications The collected data is often incomplete, especially, the label information of a large part of the data is missing. The method of finding its inherent laws from this type of data usual...

AI Technical Summary

Problems solved by technology

[0009] In order to overcome the defects of poor versatility of the existing technology on different data sets and the poor generalization performance of the obtained classifier or classification method, the present invention provides an optimization solution technology, which regards all unlabeled samples as positive samples, Taking the reliability of its label as the decision variable to be solved, an optimization model is established based on the principle of structural risk minimization, and an effective algorithm is provided to solve it.

Images