| P06-1058 | experiment process , the Naive | Bayesian modeling | is adopted for the sense clas |
| P98-1001 | 5 -RSB- . The theorem connects | Bayesian modeling | with the MDL principle in the |
| P98-1001 | MDL principle The difficulty in | Bayesian modeling | is the estimation of the prior |
| E12-1080 | presented an unsupervised dynamic | Bayesian modeling | approach to modeling speech style |
| D10-1028 | this problem using nonparametric | Bayesian modeling | , specifically adaptor grammars |
| P98-1001 | Defining the evaluation function of | Bayesian modeling | using MDL principle The difficulty |
| W12-0512 | answer scoring is built upon a | Bayesian modeling | of the process of estimating |
| W12-0512 | answers . Answer scoring through | Bayesian modeling | This method of answer scoring |
| D15-1217 | latent word in the lowest layer . | Bayesian modeling | of h-LWLM produces the following |
| D15-1217 | the number of observed words . | Bayesian modeling | of LWLM produces the generative |
| P08-1012 | Abstract We combine the strengths of | Bayesian modeling | and synchronous grammar in unsupervised |
| P98-1001 | probability . The central idea of | Bayesian modeling | is to find a compromise between |
| N09-1067 | many parameters there are . In | Bayesian modeling | , non-parametric distributions |
| W02-0214 | tification . In this case , the | Bayesian modeling | paradigm and the maximum likelihood |
| D15-1217 | , 1992 ) . Other solutions are | Bayesian modeling | ( Teh , 2006 ) and ensemble modeling |
| D14-1004 | WordNet ( in combination with | Bayesian modeling | ) is the one by O ´ S ´ |
| P98-1001 | training set and the model G , | Bayesian modeling | gives additional consideration |
| J14-3005 | 3.2 Modeling Assumptions 3.2.1 | Bayesian Modeling | . The Bayesian approach to probabilistic |
| W05-0501 | more sophisticated techniques of | Bayesian modeling | ( to replace the current mechanisms |
| H05-1032 | in a principled manner through | Bayesian modeling | , and also demonstrated how the |
