Training ν -Support Vector Classifiers: Theory and Algorithms
Citations
40,826 citations
Cites background from "Training ν -Support Vector Classifi..."
...The decision function is l t sgn yiaiK(xi,x) +b . i=1 It is shown that eT a =. can be replaced by eT a =. [Crisp and Burges 2000; Chang and Lin 2001]....
[...]
...In (Crisp and Burges, 2000; Chang and Lin, 2001), it has been shown that eα ≥ ν can be replaced by eα = ν....
[...]
10,696 citations
2,737 citations
1,586 citations
Cites background or methods from "Training ν -Support Vector Classifi..."
...The results also apply if Q is a bounded function of C since theorem 5 of Chang and Lin (2001b) holds for this case....
[...]
...function of C since Theorem 5 of ( Chang and Lin 2001b ) holds for this case....
[...]
...( Chang and Lin 2001b, Lemma 4)), A ≤ 0. However, A cannot be negative as otherwise Pl i=1 ξi goes to −∞ as C increases....
[...]
...By corollary 1 of Chang and Lin (2001b), we get linear separability in z-space....
[...]
...It can be shown (see the proof of Theorem 5 in ( Chang and Lin 2001b ) for...
[...]
611 citations
References
40,147 citations
26,531 citations
"Training ν -Support Vector Classifi..." refers methods in this paper
...This formulation is different from the original C-SVM (Vapnik, 1998): (PC) min 1 2 wTw + C l∑ i=1 ξi yi(wTφ(xi) + b) ≥ 1 − ξi, ξi ≥ 0, i = 1, . . . , l. (1.2) Neural Computation 13, 2119–2147 (2001) c© 2001 Massachusetts Institute of Technology In equation 1.2, a parameter C is used to penalize…...
[...]
...This formulation is different from the original C-SVM (Vapnik, 1998):...
[...]
12,940 citations
"Training ν -Support Vector Classifi..." refers background in this paper
...Problems adult4 and web7 are compiled by Platt (1998) from the UCI Machine Learning Repository (Blake & Merz, 1998; Murphy & Aha 1994)....
[...]
5,506 citations
5,350 citations
"Training ν -Support Vector Classifi..." refers background or methods in this paper
...Currently major methods of solving large DC (for example, decomposition methods (Osuna, Freund, & Girosi, 1997; Joachims, 1998; Platt, 1998; Saunders et al., 1998) and the method of nearest points (Keerthi, Shevade, & Murthy, 2000)) use the simple structure of constraints....
[...]
...Problems adult4 and web7 are compiled by Platt (1998) from the UCI Machine Learning Repository (Blake & Merz, 1998; Murphy & Aha 1994)....
[...]