An Epsilon Hierarchical Fuzzy Twin Support Vector Regression

TL;DR

This paper introduces a hierarchical fuzzy support vector regression model that incorporates epsilon-insensitive functions and fuzzy numbers to better handle uncertainty in forecasting, demonstrating improved generalization and efficiency.

Contribution

It proposes a novel epsilon hierarchical fuzzy twin support vector regression model combining fuzzy logic and hierarchical structure for enhanced forecasting accuracy.

Findings

Achieves superior generalization performance on synthetic and real datasets.

Reduces training time compared to existing methods.

Effectively manages uncertainty in forecasting problems.

Abstract

The research presents epsilon hierarchical fuzzy twin support vector regression based on epsilon fuzzy twin support vector regression and epsilon twin support vector regression. Epsilon FTSVR is achieved by incorporating trapezoidal fuzzy numbers to epsilon TSVR which takes care of uncertainty existing in forecasting problems. Epsilon FTSVR determines a pair of epsilon insensitive proximal functions by solving two related quadratic programming problems. The structural risk minimization principle is implemented by introducing regularization term in primal problems of epsilon FTSVR. This yields dual stable positive definite problems which improves regression performance. Epsilon FTSVR is then reformulated as epsilon HFTSVR consisting of a set of hierarchical layers each containing epsilon FTSVR. Experimental results on both synthetic and real datasets reveal that epsilon HFTSVR has remarkable generalization performance with minimum training time.