1dspam_train(1) DSPAM dspam_train(1)
2
3
4
6 dspam_train - train a corpus of mail
7
8
10 dspam_train [username] [--client] [-i index|spam_corpus nonspam_corpus]
11
12
14 dspam_train is used to train and test a corpus of mail (in maildir or
15 MBOX format). This tool will present each message to DSPAM for a clas‐
16 sification and then retrain only if the message was incorrect. This
17 provides close to real-world training and should be used to build pre‐
18 trained databases. Upon execution, the tool will automatically deter‐
19 mine the ratio of spam:nonspam and train based on that ratio to ensure
20 both corpora are trained consecutively. This tool can also be used as a
21 test jig to measure the efficiency and accuracy of a particular corpus
22 against DSPAM in a given configuration.
23
24
26 --client
27 If specified, DSPAM is used in client-server mode.
28
29
30 username
31 Specifies the user to train, if omitted the current user
32 name is used.
33
34
35 -i index
36 Use a index file instead of the usual spam_corpus and
37 nonspam_corpus.
38
39 index : Path to the index file having the following for‐
40 mat per line:
41 [class] [path to message]
42
43
44 spam_corpus
45 Specifies either the pathname to the directory containing
46 the corpus of spam, with each in a separate file (e.g.
47 maildir format) or a path to the mailbox in the tradi‐
48 tional Unix MBOX format.
49
50
51 nonspam_corpus
52 Specifies either the pathname to the directory containing
53 the corpus of nonspam with each message in a separate
54 file or a path to the mailbox in the traditional Unix
55 MBOX format.
56
57
59 0 Operation was successful.
60 other Operation resulted in an error.
61
62
64 Copyright © 2002-2011 DSPAM Project
65 All rights reserved.
66
67 For more information, see http://dspam.sourceforge.net.
68
69
71 dspam(1), dspam_admin(1), dspam_clean(1), dspam_crc(1),
72 dspam_dump(1), dspam_logrotate(1), dspam_merge(1),
73 dspam_stats(1)
74
75
76
77DSPAM Apr 17, 2010 dspam_train(1)