A note on Platt’s probabilistic outputs for support vector machines
First Online: 08 August 2007 Received: 17 February 2006 Revised: 07 May 2007 Accepted: 25 June 2007 DOI:
Cite this article as: Lin, H., Lin, C. & Weng, R.C. Mach Learn (2007) 68: 267. doi:10.1007/s10994-007-5018-6 Abstract
Platt’s probabilistic outputs for Support Vector Machines (Platt, J. in Smola, A., et al. (eds.) Advances in large margin classifiers. Cambridge,
) has been popular for applications that require posterior class probabilities. In this note, we propose an improved algorithm that theoretically converges and avoids numerical difficulties. A simple and ready-to-use pseudo code is included.
2000 Keywords Support vector machine Posterior probability Download to read the full article text References
Chang, C.-C., & Lin, C.-J. (2001). LIBSVM: a library for support vector machines. Software available at
Fletcher, R. (1987).
Practical methods of optimization
. New York: Wiley.
Goldberg, D. (1991). What every computer scientist should know about floating-point arithmetic.
ACM Computing Surveys
Moré, J. J. (1978). The Levenberg–Marquardt algorithm: implementation and theory. In G. Watson (Ed.),
(pp. 105–116). Berlin: Springer.
Nash, S. G., & Sofer, A. (1996).
Linear and nonlinear programming. New York: McGraw–Hill.
Newman, D. J., Hettich, S., Blake, C. L., & Merz, C. J. (1998).
UCI repository of machine learning databases (Technical report). Department of Information and Computer Sciences, University of California, Irvine.
Nocedal, J., & Wright, S. J. (1999).
. New York: Springer.
Platt, J. (2000). Probabilistic outputs for support vector machines and comparison to regularized likelihood methods. In A. Smola, P. Bartlett, B. Schölkopf, & D. Schuurmans (Eds.),
Advances in large margin classifiers. Cambridge: MIT Press.
Press, W. H., Flannery, B. P., Teukolsky, S. A., & Vetterling, W. T. (1992).
Numerical recipes: the art of scientific computing (2nd ed.). Cambridge: Cambridge University Press.
© Springer Science+Business Media, LLC 2007