A non-parametric semi-supervised discretization method

作者：Bondu Alexis^*; Boulle Marc; Lemaire Vincent

来源：Knowledge and Information Systems, 2010, 24(1): 35-57.

DOI：10.1007/s10115-009-0230-2

• Bayesian
• Semi-supervised
• Discretization

Semi-supervised classification methods aim to exploit labeled and unlabeled examples to train a predictive model. Most of these approaches make assumptions on the distribution of classes. This article first proposes a new semi-supervised discretization method, which adopts very low informative prior on data. This method discretizes the numerical domain of a continuous input variable, while keeping the information relative to the prediction of classes. Then, an in-depth comparison of this semi-supervised method with the original supervised MODL approach is presented. We demonstrate that the semi-supervised approach is asymptotically equivalent to the supervised approach, improved with a post-optimization of the intervals bounds location.

• 出版日期2010-7