2011 | OriginalPaper | Chapter
Incorporating a Priori Knowledge from Detractor Points into Support Vector Classification
Author : Marcin Orchel
Published in: Adaptive and Natural Computing Algorithms
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
In this article, we extend the idea of a priori knowledge in the form of detractor points presented recently for Support Vector Classification. We show that detractor points can belong to the new type of support vectors – training samples which lie outside a margin bounded region. We present the new application for a priori knowledge from detractor points – improving generalization performance of Support Vector Classification while reducing a complexity of a model by removing a bunch of support vectors. The experiments show that indeed the new type of a priori knowledge improves generalization performance of reduced models. The tests were performed on selected classification data sets, and on stock price data from public domain repositories.