[Search-l] "dependson having something to workwith"

[Jacob Kristian Andresen]

I think it could be useful to have a framework to compare various
approaches/algorithms.

So to answer questions like is approach A "better" than approach B?

This would involve thinking up a way to collect user queries and which
results the users expect the queries to return. Then we could do som
Precision/Recall measurements on the different approaches/algorithms,


Jacob


On 11/05/07, jer <jeremie at jabber.org> wrote:
>
> > what are you doing on the servers? Is there a url for your search
> > engine? Would it make sense for me to have an account?
>
> I don't have a lot of specifics yet, but I'm in the process of
> setting up Xen and my goal is to create a number of "playgrounds" for
> anyone that wants to experiment (on some decent hardware and free
> bandwidth) with YaCy, Nutch, or any other open source search tool.
> I'd really like this to be for those that don't have access to such
> resources and need an area to play/learn.
>
> As part of this I want to start gathering local copies of some useful
> shared data repositories, snapshots of DMOZ, Wikipedia, etc.
>
> I'm also going to set up a shared source management / development
> area for some upcoming projects.
>
> Any other suggestions?
>
> Jer
>
> _______________________________________________
> Search-l mailing list
> Search-l at wikia.com
> http://lists.wikia.com/mailman/listinfo/search-l
> Change options or unsubscribe:
> http://lists.wikia.com/mailman/options/search-l
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.wikia.com/pipermail/search-l/attachments/20070511/4d24289b/attachment.html