[Topic-models] Regarding LDA

Sinead Williamson saw56 at cam.ac.uk
Wed Nov 5 03:59:17 EST 2008


Dear Veena,

\alpha is 1 x K, not M x K - it is common to all documents.

Regards,

Sinead

On Wed, Nov 5, 2008 at 7:08 AM, Veena T <veenat2005 at gmail.com> wrote:

> Hello,
>
> This regarding the LDA model for document corpus.
> I have read the paper and found that at the end of learning, we have to
> estimate two parametrs:
> 1. beta
> 2. alpha
> in addition to the variational parameters.
>
> If I am correct,
> 1)we have to estimate KxV number of (beta) parameters where K corresponds
> to the number of factors and V corresponds to the cardinality of the
> multinomial variable (w)m.
> 2) We have to estimate MxK number of (alpha) parameters where M corresponds
> to the number of documents considered.
>
> Please tell me whether my understanding about the number of paremeters to
> be estimated is correct or not?
>
> regards
> veena
>
>
> --
> Veena Srinivas,
> PhD scholar,
> Speech and Vision Lab
> IIT Madras, Chennai-36
> Tamil Nadu
>
> _______________________________________________
> Topic-models mailing list
> Topic-models at lists.cs.princeton.edu
> https://lists.cs.princeton.edu/mailman/listinfo/topic-models
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/topic-models/attachments/20081105/d677750f/attachment.htm>


More information about the Topic-models mailing list