[Xapian-devel] GSoC 2011 Weighting Schemes

[wuwenjin]

I skim some related research papers, DPH, the parameter free model derived
from DfR,  achieves good performance.
Thus, DPH model should be in.
*
*
*
>
> BB2 Bernoulli-Einstein model with Bernoulli after-effect and normalisation
> 2.

IFB2 Inverse Term Frequency model with Bernoulli after-effect and
> normalisation 2.

*
*
*
*These two models don't condition on *"classic ad-hoc tasks" or "tasks that
require early precision". but I have no idea about their effectiveness.

DFR model also be used for query expansion. I will try to read
 Enquire::get_eset().
*
*
*Wenjin Wu*




2011/3/29 Olly Betts <olly at survex.com>

> On Mon, Mar 28, 2011 at 08:26:05PM +0800, wuwenjin wrote:
> > As described in http://terrier.org/docs/current/dfr_description.html,
> there
> > are many DFR models, which models will to be implemented in Xapian.
>
> DPH is apparently very effective, and it's parameter free (not having
> to tune parameters to get the best results is good), so it would
> definitely be good to have that one.
>
> I'm not sure which others are the most interesting.  Some models are
> better for some situations than others - the page you link to mentions
> "classic ad-hoc tasks" and "tasks that require early precision", but I
> am not sure which amongst those is the best option.
>
> I'd suggest picking a representative selection, and aiming to do those.
>
> There's also scope for implementing DfR query expansion in
> Enquire::get_eset() if that interests you.  That's probably more
> involved since the weights used there aren't pluggable already, so might
> be a good "if there's time at the end" thing to look at, once you're
> familiar with the query weighting.
>
> Cheers,
>     Olly
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xapian.org/pipermail/xapian-devel/attachments/20110329/80d2d573/attachment.htm>