Bayesian: Difference between revisions

25 bytes added ,  27 August 2005
no edit summary
No edit summary
No edit summary
Line 1: Line 1:
This will be a page where I want to start working on new updated Bayesian filtering functionality. My very first plan is to create a new component, more flexible (for example multistate - not only bitwise spam/nospam results but also usable for mail classification, labeling etc.). Then polish it, update Thunderbird to use it, move it to Toolkit and use it for Firefox (links prefetching, intelligent bookmarks etc.).
This will be a page where I want to start working on new updated Bayesian filtering functionality. My very first plan is to create a new component, more flexible (for example multistate - not only bitwise spam/nospam results but also usable for mail classification, labeling etc.), using recent alghoritms. Then polish it, update Thunderbird to use it, move it to Toolkit and use it for Firefox (links prefetching, intelligent bookmarks etc.).


I started reading current implementation, and planning new classes. I will need a lot of help to properly implement this and to make it work good in Mozilla way.
I started reading current implementation, and planning new classes. I will need a lot of help to properly implement this and to make it work good in Mozilla way.
Confirmed users
1,016

edits