论文信息 - Support vector machines for the classification of western handwritten capitals

Support vector machines for the classification of western handwritten capitals

In thispaper, new te hniques are presented using Support Ve tor Ma hines(SVMs) for multi- lass lassi(cid:12) ation problems. The issue of de omposing a N- lass las-si(cid:12) ation problem into a set of 2- lass lassi(cid:12) ation questions is dis ussed. In parti ular, the te hnique for normalizing the outputs of several SVMs is presented. Based on these te hniques, support ve tor lassi(cid:12)ers for the re ognition of West- ern handwritten apitals are realized. Comparisons to several other lassi(cid:12) ation methods are also presented.

[1] J. Movellan. Tutorial on Hidden Markov Models , 2006 .

[2] Christopher M. Bishop Geoffrey Hinton,et al. Neural Networks For Pattern Recognition , 2005 .

[3] Louis Vuurpijl,et al. Two tree-formation methods for fast pattern search using nearest-neighbour and nearest-centroid matching , 2000 .

[4] M. Weinfeld,et al. TREE SEARCH TECHNIQUE FOR THE OPTIMIZATION OF THE k NEAREST NEIGHBORS ALGORITHM , 1999 .

[5] K. Schittkowski,et al. NONLINEAR PROGRAMMING , 2022 .