We present a new method for information retrievalusing hidden Markov models (HMMs). We develop a general framework for incorporating multiple word generation mechanisms within the...
Existing data cleaning methods work on the basis of computing the degree of similarity between nearby records in a sorted database. High recall is achieved by accepting records wi...
In this paper, an original method named GNG-T, extended from GNG-U algorithm by [1] is presented. The method performs continuously vector quantization over a distribution that chan...
We present a learning framework for structured support vector models in which boosting and bagging methods are used to construct ensemble models. We also propose a selection metho...
This paper investigates the effect of Kernel Principal Component Analysis (KPCA) within the classification framework, essentially the regularization properties of this dimensional...