Title :
Shrinking the Tube: A New Support Vector Regression Algorithm with Parametric Insensitive Model
Author_Institution :
Nat. Kaohsiung Univ. of Appl. Sci., Kaohsiung
Abstract :
A new algorithm for support vector regression is described. For a priori chosen v, it automatically adjusts a flexible tube of arbitrary shape and minimal radius to include the data such that at most a fraction v of the data points lie outside. Moreover, it is shown how to use parametric tube shapes with non-constant radius. The algorithm is analysed theoretically and experimentally.
Keywords :
learning (artificial intelligence); regression analysis; support vector machines; parametric insensitive model; parametric tube shape; statistical learning; support vector machine; support vector regression algorithm; Algorithm design and analysis; Approximation algorithms; Cybernetics; Electronic mail; Information management; Machine learning; Machine learning algorithms; Shape; Statistical learning; Support vector machines; Insensitive model; Interval regression; Support vector machines; Support vector regression;
Conference_Titel :
Machine Learning and Cybernetics, 2007 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-0973-0
Electronic_ISBN :
978-1-4244-0973-0
DOI :
10.1109/ICMLC.2007.4370453
