[Topic-models] Running LDA on "20 newsgroups"

Hesam Alian hesam.amoualian at imag.fr
Fri Nov 4 07:56:22 EDT 2016


Hi

I think, there isn't such of this version but you can easily make it using this python code :
https://github.com/JoKnopp/text2ldac <https://github.com/JoKnopp/text2ldac>
you either need to run the code for each folder of the raw text or put all together in one folder and run the code for them.

Best
H
> On 4 Nov 2016, at 19:43, Rabe'e Ravanifard <ravanifard at gmail.com> wrote:
> 
> Dear All,
> I want to run LDA on "20 newsgroups" dataset. Is there any versions
> which its data format is acceptable by LDA? (each document has been
> represented as
> [number of uniqe terms] [term1]:[count] [term_2]:[count] ...  [term_N]:[count] )
> 
> I would be grateful if you could help me.
> 
> Best Regards,
> Rabeh
> _______________________________________________
> 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/20161104/134a3394/attachment.html>


More information about the Topic-models mailing list