W04-1907 information loss in the model due to discretisation . Also , if the annotation is
P06-2085 for feature selection and PKI discretisation . This model achieves a wf -
P08-1073 of the multimodal function as discretisation boundaries . Previous work on
P06-2085 or MDL discretised data . MDL discretisation reduces our range of feature
S15-2022 - cretisation . Regression by discretisation follows precisely the same methodology
P06-2085 . All parameters , except for discretisation methods have a significant impact
P06-2085 quality of our data for ML . First , discretisation methods take feature distributions
P06-2085 Discretising Numeric Features Global discretisation methods divide all continuous
P08-1073 Lemon , 2008a ) . 3.6 State space discretisation We use linear function approximation
P06-2085 Discretisation Feature selection and discretisation influence one - another , i.e.
P06-2085 context features . • For discretisation methods we found that the classifiers
N07-4014 . The system uses a grid-based discretisation of the state space and online
P06-2085 comparing the group means for models , discretisation , and features selection methods
P06-2085 all instances . However , MDL discretisation can not replace proper feature
P06-2085 the effects of models , feature discretisation and selection on performance
P06-2085 experiments we use implementations of discretisation and feature selection methods
N07-2038 The HIS model uses a grid-based discretisation of the continuous state space
P06-2085 behaved uniformly , using feature discretisation methods and feature selection
W12-3108 accuracies appear only with a discretisation interval of 1.00 , which though
P06-2085 accuracy ( R2 = .181 ) whereas discretisation methods only contribute 0.4 %
hide detail