| 1 | .\" $Id: dspam_train.1,v 1.10 2011/06/28 00:13:48 sbajic Exp $ |
|---|
| 2 | .\" -*- nroff -*- |
|---|
| 3 | .\" |
|---|
| 4 | .\" dspam_train3.9 |
|---|
| 5 | .\" |
|---|
| 6 | .\" Authors: Jonathan A. Zdziarski <jonathan@nuclearelephant.com> |
|---|
| 7 | .\" Stevan Bajic <stevan@bajic.ch> |
|---|
| 8 | .\" |
|---|
| 9 | .\" Copyright (C) 2002-2012 DSPAM Project |
|---|
| 10 | .\" All rights reserved |
|---|
| 11 | .\" |
|---|
| 12 | .TH dspam_train 1 "Apr 17, 2010" "DSPAM" "DSPAM" |
|---|
| 13 | |
|---|
| 14 | .SH NAME |
|---|
| 15 | dspam_train \- train a corpus of mail |
|---|
| 16 | |
|---|
| 17 | .SH SYNOPSIS |
|---|
| 18 | .na |
|---|
| 19 | .B dspam_train |
|---|
| 20 | [\c |
|---|
| 21 | .BI username\fR\c |
|---|
| 22 | ] |
|---|
| 23 | [\c |
|---|
| 24 | .BI \--client\fR\c |
|---|
| 25 | ] |
|---|
| 26 | [\c |
|---|
| 27 | .BI \-i\ \fR\c |
|---|
| 28 | index|\c |
|---|
| 29 | .BI spam_corpus\fR\c |
|---|
| 30 | \ \c |
|---|
| 31 | .BI nonspam_corpus\fR\c |
|---|
| 32 | ] |
|---|
| 33 | |
|---|
| 34 | .ad |
|---|
| 35 | .SH DESCRIPTION |
|---|
| 36 | .LP |
|---|
| 37 | .B dspam_train |
|---|
| 38 | is used to train and test a corpus of mail (in maildir or MBOX format). This |
|---|
| 39 | tool will present each message to DSPAM for a classification and then |
|---|
| 40 | retrain only if the message was incorrect. This provides close to real\-world |
|---|
| 41 | training and should be used to build pretrained databases. Upon execution, |
|---|
| 42 | the tool will automatically determine the ratio of spam:nonspam and train |
|---|
| 43 | based on that ratio to ensure both corpora are trained consecutively. This |
|---|
| 44 | tool can also be used as a test jig to measure the efficiency and accuracy |
|---|
| 45 | of a particular corpus against DSPAM in a given configuration. |
|---|
| 46 | |
|---|
| 47 | .SH OPTIONS |
|---|
| 48 | .LP |
|---|
| 49 | .ne 3 |
|---|
| 50 | .TP |
|---|
| 51 | |
|---|
| 52 | .ne 3 |
|---|
| 53 | .TP |
|---|
| 54 | .BI \--client\c |
|---|
| 55 | If specified, DSPAM is used in client\-server mode. |
|---|
| 56 | |
|---|
| 57 | .ne 3 |
|---|
| 58 | .TP |
|---|
| 59 | .BI username\c |
|---|
| 60 | Specifies the user to train, if omitted the current user name is used. |
|---|
| 61 | |
|---|
| 62 | .ne 3 |
|---|
| 63 | .TP |
|---|
| 64 | .BI \-i\fR\ index\c |
|---|
| 65 | Use a index file instead of the usual spam_corpus and nonspam_corpus. |
|---|
| 66 | |
|---|
| 67 | .B index |
|---|
| 68 | : Path to the index file having the following format per line: |
|---|
| 69 | .br |
|---|
| 70 | [class] [path to message] |
|---|
| 71 | |
|---|
| 72 | .ne 3 |
|---|
| 73 | .TP |
|---|
| 74 | .BI spam_corpus\c |
|---|
| 75 | Specifies either the pathname to the directory containing the corpus of spam, |
|---|
| 76 | with each in a separate file (e.g. maildir format) or a path to the mailbox in |
|---|
| 77 | the traditional Unix MBOX format. |
|---|
| 78 | |
|---|
| 79 | .ne 3 |
|---|
| 80 | .TP |
|---|
| 81 | .BI nonspam_corpus\c |
|---|
| 82 | Specifies either the pathname to the directory containing the corpus of |
|---|
| 83 | nonspam with each message in a separate file or a path to the mailbox in the |
|---|
| 84 | traditional Unix MBOX format. |
|---|
| 85 | |
|---|
| 86 | .SH EXIT VALUE |
|---|
| 87 | .LP |
|---|
| 88 | .ne 3 |
|---|
| 89 | .PD 0 |
|---|
| 90 | .TP |
|---|
| 91 | .B 0 |
|---|
| 92 | Operation was successful. |
|---|
| 93 | .ne 3 |
|---|
| 94 | .TP |
|---|
| 95 | .B other |
|---|
| 96 | Operation resulted in an error. |
|---|
| 97 | .PD |
|---|
| 98 | |
|---|
| 99 | .SH COPYRIGHT |
|---|
| 100 | Copyright \(co 2002\-2012 DSPAM Project |
|---|
| 101 | .br |
|---|
| 102 | All rights reserved. |
|---|
| 103 | .br |
|---|
| 104 | |
|---|
| 105 | For more information, see http://dspam.sourceforge.net. |
|---|
| 106 | |
|---|
| 107 | .SH SEE ALSO |
|---|
| 108 | .BR dspam (1), |
|---|
| 109 | .BR dspam_admin (1), |
|---|
| 110 | .BR dspam_clean (1), |
|---|
| 111 | .BR dspam_crc (1), |
|---|
| 112 | .BR dspam_dump (1), |
|---|
| 113 | .BR dspam_logrotate (1), |
|---|
| 114 | .BR dspam_merge (1), |
|---|
| 115 | .BR dspam_stats (1) |
|---|
