Talk:Platt scaling

I'm drafting a rewrite of this page to rename it (probably to "Calibration (machine learning)"), to talk more generally about both univariate scaling of classifier outputs and other calibration methods, and to include citations to earlier work (including my own) on the subject. I've discussed these changes with John Platt, who agrees that the entry should be renamed. — Preceding unsigned comment added by Daviddlewis (talk • contribs)

Please be aware that the article Probabilistic classification contains a section "Probability calibration" that covers the broader topic. Expanding that section might be a better idea than starting from this page. QVVERTYVS (hm?) 13:24, 3 August 2014 (UTC)

Qwertyus - OK, excellent, so Probabilistic Classification covers the range of approaches for creating calibrated classifiers. The choice would then be whether the Platt scaling page should become one on all univariate rescaling, or just on univariate logistic regression rescaling. Thoughts? Dave Daviddlewis (talk) 14:39, 3 August 2014 (UTC) [copied here by QVVERTYVS (hm?) 07:47, 4 August 2014 (UTC)]

I don't really mind. I think Platt scaling is a "classic" that deserves its own page, but if other methods are really very similar we can broaden the scope of this page as well and make it the {{main}} for the "Probability calibration" section.

Do you have a literature list? Here's what we're citing so far:

• Hastie, Trevor; Tibshirani, Robert (1998). "Classification by pairwise coupling". The Annals of Statistics. 26 (2): 451–471. doi:10.1214/aos/1028144844. Zbl 0932.62071. CiteSeerX: 10.1.1.46.6032.
• Platt, John (1999). "Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods" (PDF). Advances in large margin classifiers. 10 (3): 61–74.
• Olivier Chapelle; Vladimir Vapnik; Olivier Bousquet; Sayan Mukherjee (2002). "Choosing multiple parameters for support vector machines" (PDF). Machine Learning. 46: 131–159.
• Gebel, Martin; Weihs, Claus (2008). "Data Analysis, Machine Learning and Applications". Studies in Classification, Data Analysis, and Knowledge Organization: 29. doi:10.1007/978-3-540-78246-9_4. ISBN 978-3-540-78239-1. {{cite journal}}: |chapter= ignored (help); Cite journal requires |journal= (help)
• Niculescu-Mizil, Alexandru; Caruana, Rich (2005). Predicting good probabilities with supervised learning (PDF). ICML.
• Lin, Hsuan-Tien; Lin, Chih-Jen; Weng, Ruby C. (2007). "A note on Platt's probabilistic outputs for support vector machines" (PDF). Machine Learning. 68 (3): 267–276.

QVVERTYVS (hm?) 07:55, 4 August 2014 (UTC)

Vladimir Vapnik

"replacing an earlier method by Vapnik"

Please write out the entire name. https://en.wikipedia.org/wiki/Vladimir_Vapnik — Preceding unsigned comment added by 140.163.254.158 (talk) 20:38, 2 October 2018 (UTC)