Jorge  Civera  and Alfons  Juan.   Multinomial  Mixture Modelling  for
Bilingual Text Classification.   In Proc. of the 6th  Int. Workshop on
Pattern Recognition in Information  Systems (PRIS 2006), pages 93-103,
Paphos (Cyprus), May 2006.

Abstract.   Mixture  modelling  of  class-conditional densities  is  a
standard pattern classification technique. In text classification, the
use  of  class-conditional  multinomial  mixtures  can be  seen  as  a
generalisation  of  the  Naive  Bayes  text  classifier  relaxing  its
(class-conditional feature) independence assumption. In this paper, we
describe  and  compare  several  extensions of  the  class-conditional
multinomial mixture-based text classifier for bilingual texts.