GSOC Proposal

[Mehak Goyal]

Hey!

Sorry for reverting too late. I got stuck with the university mid-semester
exams. Just completed them the previous week. I realize I have few days to
get my proposal in line.

I have built the code for Xapian on my Linux System. In line with my
intended project, Learning to rank ClickStream Data Mining, I have reviewed
the resources available with the project.

My understanding of the project is that we are trying to train LeTor
through User clicks without the user having to train explicitly with
relevance judgment. To capture the relevance judgments, currently, we have
deployed the counting algorithm based DBN model and we wish to further
improve upon it using Expectation Maximization and Forward-Backward
algorithms and also include other models (Dependent Click Model and Intent
Aware Model) for comparison to the user.

I have reviewed the available literature. I understand the working of Letor
and DBN and why it is preferred over the conventional methods: it includes
the effect of the other documents in the search results and also the effect
of position and perceived relevance. Also, I underwent the previous GSoC
project in this direction and from where I need to start off.

I understand I have a considerable grasp of the concepts and I wish to know
how to proceed further to be able to draft the proposal. What is expected
of me at this stage when I am done with building Xapian and reviewing the
available, relevant literature.

I am very much interested in working with Xapian given my background and my
interest. A prompt response shall be appreciated.

Regards
Mehak

On Thu, Feb 15, 2018 at 3:20 AM, Olly Betts <olly at survex.com> wrote:

> Hi Mehak,
>
> On Wed, Feb 14, 2018 at 02:57:35PM +0530, Mehak Goyal wrote:
> > I am exploring the project specific page of Letor.
> >
> > I would like to know more details that could help me get started on the
> > project.
>
> That's a very open-ended and generic question and so hard to really
> answer usefully.
>
> There's quite a lot of detail on the project idea page - what are you
> specifically wanting to know about that isn't covered there?
>
> > Also, help with how to go about the project proposal shall be
> appreciated.
>
> We recommend you start by working through the guide we put together for
> this purpose:
>
> https://trac.xapian.org/wiki/GSoC%20Guide
>
> That helps you obtain the code, build it, and get to grips with how to
> modify it.
>
> Cheers,
>     Olly
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xapian.org/pipermail/xapian-devel/attachments/20180321/181c6321/attachment.html>