Manipulation of prior probabilities in support vector classification
Authors:
- G.C. CawleyGCG.C. Cawley
- N.L.C. TalbotNTN.L.C. Talbot
- 1 July 2001
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
Asymmetric margin error costs for positive and negative examples are often cited as an efficient heuristic compensating for unrepresentative priors in training support vector classifiers. In this paper we show that this heuristic is well justified via simple re-sampling ideas applied to the dual Lagrangian defining the 1-norm soft-margin support vector machine. This observation also provides a simple expression for the asymptotically optimal ratio of margin error penalties, eliminating the need for the trial-and-error experimentation normally encountered. This method allows the use of a smaller, balanced training data set in problems characterised by widely disparate prior probabilities, reducing the training time. The usefulness of this method is then demonstrated on a real world benchmark probleKeywords
This publication has 11 references indexed in Scilit:
- An Introduction to Support Vector Machines and Other Kernel-based Learning Methods2000
- Comparative accuracies of artificial neural networks and discriminant analysis in predicting forest cover types from cartographic variablesComputers and Electronics in Agriculture, 1999
- A Tutorial on Support Vector Machines for Pattern RecognitionData Mining and Knowledge Discovery, 1998
- Neural Network Classification and Prior Class Probabilities1998
- Support-vector networksMachine Learning, 1995
- The Nature of Statistical Learning Theory1995
- A training algorithm for optimal margin classifiers1992
- Neural Network Classifiers Estimate Bayesian a posteriori ProbabilitiesNeural Computation, 1991
- Exploiting prior knowledge in network optimization: an illustration from medical prognosisNetwork: Computation in Neural Systems, 1990
- Single-layer learning revisited: a stepwise procedure for building and training a neural network1990