I think that there is some confusion here. The latent factors whether they number 5 or 100 are really not the same as topics.
Secondly, documents *can* be about more than one subject so mixtures are good to have a representation for. Thirdly, not all concepts are totally distinct. Most if not all are related to some other topics. This means a representation that allows for this gradated similarity is good. On Sat, Oct 23, 2010 at 1:58 AM, Sid <[email protected]> wrote: > So my question is that if there are a lot of topics bunched in, like you > suggested i should do with 5-100; That may give me multiple concepts for > the > same documents; concepts that are actually conceptually different and not > all being representative of the query document. > Is my interpretation here correct? >
