jm + spam + searching   1

feedback loop n-gram analyzer
'a simple parser of ARF compliant FBL complaints, which normalizes the email complaints and generates a 6-tuple n-gram version of the message. These n-grams are stored in a Redis database, keyed by the file in which they can be found. An inverse index also exists that allow you to find all messages containing a particular n-gram word.'
anti-spam  spam  fbl  feedback  filtering  n-grams  similarity  hashing  redis  searching 
september 2011 by jm

Copy this bookmark: