I'm using R package stm for topic modelling.
But I found that every time I run stm the Top Topics results are different.
First_STM <- stm(documents = out$documents, vocab = out$vocab, K = 10, prevalence = ~ rating, max.em.its = 75, data = out$meta, init.type = "Spectral", verbose = TRUE)
The result of First_STM is like that: enter image description here
When I use the SAME code and run stm for another time:
Second_STM <- stm(documents = out$documents, vocab = out$vocab, K = 10, prevalence = ~ rating, max.em.its = 75, data = out$meta, init.type = "Spectral", verbose = TRUE)
The result of Second_STM is like that: enter image description here
My question is: why I use the SAME set of code, but the results are different.
I have tried running stm in R for Top Topics with the same code but results are different. I expect that the results would be the same if I use the same code.