: >> but assuming "clean data" with no mis-spellings, scoring "rare" terms : higher seems like the ideal behavior : : Exact matched of a term should have a higher ranking then fuzzy matched : terms.. at least that's the expected behaviour in my situation although : i think it seems the most common behaviour.. We could also make this : optionally, so that we don't affect the current scoring algorithm..
Unless I'm *really* confused by what the existing patch does, then what you are proposing would be a third, completley different, scoring approach -- as i understand the patch, less-fuzzy matches don't score better, it's just that term with the biggest IDF sets the IDF for all other terms. -Hoss --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
