 Cybernetics and Systems Analysis /  Issue (2020, 56 (3))
Zhurbenko N.G.
Linear classifier and projection onto a polytope An algorithm for constructing binary linear classifiers is considered. Objects of recognition are presented by points of an n-dimensional Euclidean space. The algorithm is based on solving the problem of projecting zero onto the convex hull of a finite number of points of Euclidean space. © 2020, Springer Science+Business Media, LLC, part of Springer Nature. Keywords: convex hull, linear classifier, penalty functions, projection on polytope, Computer science, Cybernetics, Binary linear classifier, Convex hull, Euclidean spaces, Finite number, Linear classifiers, Polytopes, Geometry
Cite:
Zhurbenko N.G.
(2020). Linear classifier and projection onto a polytope. Cybernetics and Systems Analysis, 56 (3), 163–170. doi: https://doi.org/10.1007/s10559-020-00264-3 http://jnas.nbuv.gov.ua/article/UJRN-0001121528 [In Russian]. |
PDF
