[Topic-models] Topic-models Digest, Vol 72, Issue 15

Halil Bisgin hxbisgin at ualr.edu
Wed Jul 25 15:33:10 EDT 2012


If you can manipulate the distributions you got at the end and obtain
P(T|w), then you can measure the K-L distance between words. Finally, you
can start returning from the closest one until the desired number of
results.

This is the way what I can think right now.

Best,
Halil


On Tue, Jul 24, 2012 at 11:00 AM, <
topic-models-request at lists.cs.princeton.edu> wrote:

> Send Topic-models mailing list submissions to
>         topic-models at lists.cs.princeton.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         https://lists.cs.princeton.edu/mailman/listinfo/topic-models
> or, via email, send a message with subject or body 'help' to
>         topic-models-request at lists.cs.princeton.edu
>
> You can reach the person managing the list at
>         topic-models-owner at lists.cs.princeton.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Topic-models digest..."
>
> Today's Topics:
>
>    1. Similarity word LDA (Francesco Lisena)
>
>
> ---------- Forwarded message ----------
> From: Francesco Lisena <francescolisena8 at gmail.com>
> To: topic-models at lists.cs.princeton.edu
> Cc:
> Date: Tue, 24 Jul 2012 03:36:50 +0200
> Subject: [Topic-models] Similarity word LDA
> Hi all...
> I am a beginner at Mallet.
> I'm working on the technique of TopicExtraction, using the class
> ParallelTopicModel.
> I have 30,000 documents and I am able to estimate my model, and print all
> kind of report....
> The model created, has these parameters:
> - alpha = 0,1;
> - beta = 0,01;
> - numTopics= 300;
> - numThreads=4;
> - numIterations= 2000;
> In addition, I use the function TokenSequenceNGrams to represent unigrams,
> bigrams and trigrams.
>
> My problem is:
> I would like to create an algorithm that given as input a set of words
> (query words), returns as output a set of related words, using my topic
> model.
>
> For example:
> Input word: java, software
> output: java_developer, eclipse, software_ engineer
>
> Any idea? There are formulas in the literature? For example a similarity
> P(w|Q) where Q is a set of query's word.????
>
> _______________________________________________
> 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/20120725/718fa5d1/attachment.html>


More information about the Topic-models mailing list