Topic Modeling in R using LDA -
i have been trying out different ways of determining number of topics in lda (in r) , have used r package ldatuning
using method gibbs sampling , not able understand meaning of different metrics like:
metrics = c("griffiths2004", "caojuan2009", "arun2010", "deveaud2014"), method = "gibbs",
can me understand different metrics? if can provide guidance here great. in advance.
have taken @ findtopicsnumber_plot() ? try:
findtopicsnumber_plot(your_result_from_tuning)
each metric have different pros , cons , may furthermore more or less applicable given specifics of data. basic overview though want parameters minimize arun , caojuan, want parameters maximize griffiths , deveaud.
take @ this: http://rpubs.com/nikita-moor/107657
Comments
Post a Comment