mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From abhishek kumar <abhishek.kumar.cs...@itbhu.ac.in>
Subject Re: Integrating browser with apache mahout !
Date Fri, 10 Jan 2014 08:48:49 GMT
Hi Tharindu Rusira,

Thanks for replying.

Yes, I'm extracting topics from text content of the page (though I'm also
working on how to incorporate meta data and links in it for better
modelling).

 Actually I'm trying to predownload those pages which are most similar to
the url's that are visited by users in a time period (context ). For this I
want to use only those url's that are in the history of user(usually
recorded by browsers) . Also I want to train my model on some server or
large database and pipeline it to client. That is why I need some web
interface or browser to work on.

Please ask if you still have any doubt.

Aks


On Fri, Jan 10, 2014 at 2:01 PM, Tharindu Rusira
<tharindurusira@gmail.com>wrote:

> On Fri, Jan 10, 2014 at 1:42 PM, abhishek kumar <
> abhishek.kumar.cse10@itbhu.ac.in> wrote:
>
> > Hi ,
> >
> Hi Abishek,
>
> >
> > I'm new to apache mahout. I'm working in topic modelling (particularly
> > LDA), I have learnt that Mahout has various modules for machine learning
> > and topic modelling
>
> Yes, Mahout has a topic modelling component that implements LDA
>  but I have not worked with it yet
> . (org.apache.mahout.clustering.lda.cvb)
> .
>
>
>
> > .I want to use it's capabilities in browsing to
> > recommend sites (or predownload desired webpages based on topics of
> url's).
> >
> Just out of curiosity, why do you want to extract topics from URLs and not
> from the content of the page?
>
> >
> > I don't have idea how it can be incorporated or attached to any
> > browser.
>
> You want a web interface, don't you?
>
>
> > Please help me in this project also if you have any suggestions it
> > will be helpful.
> >
> >
> > Aks
> >
> Regards,
>
>
>
> --
> M.P. Tharindu Rusira Kumara
>
> Department of Computer Science and Engineering,
> University of Moratuwa,
> Sri Lanka.
> +94757033733
> www.tharindu-rusira.blogspot.com
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message