Security Investigate

Support Vector Classification With Input Data Uncertainty

Download now Free registration required

Executive Summary

This paper investigates a new learning model in which the input data is corrupted with noise. This paper presents a general statistical framework to tackle this problem. Based on the statistical reasoning, the paper proposes a novel formulation of support vector classification, which allows uncertainty in input data. The paper derives an intuitive geometric interpretation of the proposed formulation, and develops algorithms to efficiently solve it. Empirical results are included to show that the newly formed method is superior to the standard SVM for problems with noisy input.

  • Format: PDF
  • Size: 103.6 KB