An approach to the selection of hyperplane for recognition algorithms

E. I. Lysenok

Faculty of Computational Mathematics and Cybernetics, Moscow State University, Moscow, 119992 Russia

Abstract: An approach to improving the classification quality of correct algorithms based on a linear combination of algorithms based on the separation of objects by hyperplanes is proposed.

Key words: classification, recognition, hyperplane, basis, feature, object.

UDC: 519.7

Received: 05.05.2010

 English version: Computational Mathematics and Mathematical Physics, 2010, 50:10, 1767–1768

Bibliographic databases:

• 
• 
•