It will be a few weeks still. I'm in a pretty good crunch at work
right now. Still, I did like your proposal and I know my first glance
at the problem is going to be pretty basic, so I am not sure what to
recommend. Perhaps it makes sense to propose it as building on an
existing (if it exists, I can't predict the future :-) ) otherwise
you will implement from scratch. I know that leaves you hanging a
little bit, but I also hate to see your existing effort go to waste.
-Grant
On Mar 31, 2008, at 9:14 AM, Vitalisova, Natallia wrote:
Grant,
In fact, I haven't got a lot of experience in machine learning
algorithms implementation, so I think Naïve Bayes will perfectly
fit. You said you may need some help in your Naïve Bayes
implementation - I believe, for the next month it would be the best
choice for me. This time, I guess, will be sufficient for me to
investigate this topic more deeply and understand how it is
connected with other topics, and maybe then implement something more
complicated (like things about which Paul said or other ones). So, I
just need some time for investigation.
Could you, for example, commit in svn what you have already done? I
would really be eager to help.
Natallia
