Uncertainty-aware twin support vector machines

Zhizheng Liang

doi:10.1016/j.patcog.2022.108706

2022-09-01

PRE

Zhizheng Liang *

张磊 (Lei Zhang)

DOI：10.1016/j.patcog.2022.108706

原文链接

原文求助

摘要

En 中文

There exist uncertain data in the real world due to some factors such as imprecise measurements and noise. Unlike deterministic data, the features of samples in uncertain data are often described by inter-val numbers or random vectors with probability density functions. In this paper we propose novel twin support vector machines (TSVMs) to handle uncertain data. In the proposed models which are referred to as uncertainty-aware TSVMs, each uncertain sample is modeled as a random vector with Gaussian distributions. To deal with the multi-dimensional integrals in the original models, we derive an inter-esting and important theorem which helps us transform the original models into the model involving one-dimensional integrals. The simplification of models makes the optimization problem tractable and the simplified models are solved by using the quasi-Newton optimization algorithm. The proposed de-cision rule allows us to classify uncertain samples with means and covariance matrices. In addition, we extend the proposed models to their kernel versions to capture the nonlinear structure of uncertain data. Experiments on a series of data sets have been performed to demonstrate that the proposed models gain better classification performance than some existing algorithms, especially for representing uncer-tain cross-plane problems. (c) 2022 Elsevier Ltd. All rights reserved.

Keyword:

Uncertain data

Twin support vector machines

Halfspaces

Kernel functions

Data classification