nameless_centurian
Gawd
- Joined
- Apr 4, 2003
- Messages
- 836
for my master's work, i have selected a topic which has been given more consideration recently: that of classifying bugs as duplicates for some percentage of actual bug duplicates. a recent paper i read filtered 8% of the duplicates. i hope to make that percentage larger, but 8% is a respectable number for first-time work, and i'm sure some organizations would love to shrink their number of bug reports to 92% of the thousands they get annually.
my idea is kind of half-baked at this point since it is very early in the research process, but what i am looking for now is this:
1) access to bug reports through a bug-tracking tool like bugzilla so i can see what i am dealing with. where can i download the open bug reports for a project like mozilla?
2) if i have to write a text parser to do what i want, so be it.. but it's nice to save time. anybody aware of any products (fairly cheap (< $500) is good. free is better. additionally, i think this parenthetical explanation looks like LISP code.) that can parse text and find consistencies in contextual use of words? this product does not have to be super-smart.any product that fits the requirement i just laid out will be considered.
thanks for your time.
my idea is kind of half-baked at this point since it is very early in the research process, but what i am looking for now is this:
1) access to bug reports through a bug-tracking tool like bugzilla so i can see what i am dealing with. where can i download the open bug reports for a project like mozilla?
2) if i have to write a text parser to do what i want, so be it.. but it's nice to save time. anybody aware of any products (fairly cheap (< $500) is good. free is better. additionally, i think this parenthetical explanation looks like LISP code.) that can parse text and find consistencies in contextual use of words? this product does not have to be super-smart.any product that fits the requirement i just laid out will be considered.
thanks for your time.