1BOGOTUNE(1) Bogofilter Reference Manual BOGOTUNE(1)
2
6 bogotune - find optimum parameter settings for bogofilter
7
9 bogotune [-v] [-c config] [-C] [-d dir] [-D] [-r value] [-T value]
10 -n okfile [[-n] okfile [...]]
11 -s spamfile [[-s] spamfile [...]] [-M file]
12 bogotune [-h]
14
16 Bogotune tries to find optimum parameter settings for bogofilter. It
17 needs at least one set each of spam and non-spam messages. The
18 production wordlist is normally used, but it can be directed to read a
19 different wordlist, or to build its own from half the supplied
20 messages.
21 In order to produce useful results, bogotune has minimum message count
23 requirements. The wordlist it uses must have at least 2,000 spam and
24 2,000 non-spam in it and the message files must contain at least 500
25 spam and 500 non-spam messages. Also, the ratio of spam to non-spam
26 should be in the range 0.2 to 5. If you direct bogotune to build its
27 own wordlist, it will use the half the input or 2000 messages
28 (whichever is larger) for the wordlist.
29 Message files may be in mbox, maildir, or MH folder or any combination.
31 Msg-count files can also be used, but not mixed with other formats.
32
34 The -h option prints the help message and exits.
35 The -v option increases the verbosity level. Level 1 displays the scan
37 output in detail instead of using a progress meter.
38 The -c filename option tells bogofilter to read the config file named.
40 The -C option prevents bogotune from reading a configuration file.
42 The -d dir option specifies the directory for the database. See the
44 ENVIRONMENT section for other directory setting options.
45 The -D option tells bogotune to build a wordlist in memory using the
47 input messages. The messages are read and divided into two groups. The
48 first group is used to build a wordlist (in ram) and the second is used
49 for tuning. To meet the minimum requirements of 2000 messages in the
50 wordlist and 500 messages for testing, when -D is used, there must be
51 2500 non-spam and 2500 spam in the input files. If there are enough
52 messages (more than 4000), they will be split evenly between wordlist
53 and testing. Otherwise, they will be split proportionately.
54 The -n option tells bogotune that the following argument is a file (or
56 folder) containing non-spam. Since version 1.0.3, multiple arguments to
57 the -n option can be given. All non-option arguments until the next -s
58 option will be treated as though they had been preceded by -n
59 The -s option tells bogotune that the following argument is a file (or
61 folder) containing spam. It can be repeated as often as necessary.
62 Since version 1.0.3, multiple arguments to the -s can be given. All
63 non-option arguments until the next -n option will be treated as though
64 they had been preceded by -s.
65 The -r value option tells bogotune to use the following parameter as
67 the robx value.
68 The -T value option tells bogotune to use the following parameter as fp
70 target value.
71 The -M file option tells bogotune to convert the file to message count
73 format. This format provides a sorted list of each message's unique
74 tokens, along with their ham and spam counts. Sorting hides the sense
75 of the messages quite effectively, thus protecting privacy. The
76 message-count format allows bogotune and bogofilter to score messages
77 quickly without needing the original token database.
78
80 Bogofilter uses a database directory, which can be set in the config
81 file. If not set there, bogofilter will use the value of
82 BOGOFILTER_DIR. Both can be overridden by the -ddir option. If none of
83 that is available, bogofilter will use directory $HOME/.bogofilter.
84
86 Bogotune is not particularly robust when presented with garbage input.
87
89 The bogofilter developer team.
90 For updates, see the bogofilter project page[1].
92
94 bogofilter(1), bogolexer(1), bogoupgrade(1), bogoutil(1)
95
97 1. the bogofilter project page
98 http://bogofilter.sourceforge.net/
99Bogofilter 10/22/2012 BOGOTUNE(1)