Function CreateCollocationTopics

Create new vocabulary and topic counts based on collocation model output

[ WPNEW,DPNEW,WONEW] = CreateCollocationTopics( C,Z,WO,DS,WS,T,MAXC) creates a new vocabulary WONEW and new sparse W x T and W x D count matrices WPNEW and DPNEW from the output of the LDACOL (LDA Collocation) model. C and Z are vectors containing route (0=topic; 1=collocation) and topic assignments WO and DS contain the word and document indices of the corpus input to the model and T is the the number of topics.

MAXC determines the maximum number of words that can be part of a collocation remove annoying chars