I consider a binary classification problem with a feature vector of high dimensionality. Spam mail filters are a popular example hereof. A Bayesian approach requires us to estimate the probability of a feature vector given the class of the object. Due to the size of the feature vector this is an unfeasible t ask. A useful approach is to split the feature space into several (conditionally) independent subspaces. This results in a new problem, namely how to find the " best" subdivision. In this paper I consider a weighing approach that will perform (asymptotically) as good as the best subdivision and still has a manageable complexity
|Title of host publication||Proceedings of the 29th Symposium on Information Theory in the Benelux, May 29-30, 2008, Leuven, Belgium|
|Editors||L. Perre, Van der, A. Dejonghe, V. Ramon|
|Place of Publication||Leuven|
|Publication status||Published - 2008|
Tjalkens, T. J. (2008). Four model classes for efficient Bayesian selection. In L. Perre, Van der, A. Dejonghe, & V. Ramon (Eds.), Proceedings of the 29th Symposium on Information Theory in the Benelux, May 29-30, 2008, Leuven, Belgium (pp. 121-128). IMEC.