calc_imi | R Documentation |
The actual computation of IMI (instantaneous mutual information) values for
imi_topic
is done by this routine. If you are calling
imi_topic
repeatedly for a given topic, you can gain some speed by
pre-calculating the inputs and supplying them directly to this function.
calc_imi(doc_topics_k, term_doc_k)
doc_topics_k |
vector of counts, one for each document, of words assigned to topic k |
term_doc_k |
matrix of counts, where element i,j is the number of words of type i in document (or document-group) j is assigned to topic k |
a vector of scores, one for each row of term_doc_k
I have referred to a helpful note by David Mimno here: https://lists.cs.princeton.edu/pipermail/topic-models/2012-March/001779.html.
imi_topic
which calls this; see under "Details" there
for more detail on IMI
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.