83:
124:
64:
54:
117:
143:
153:
110:
20:
82:
148:
60:
34:
to determine the weights for each of those features in terms of their predictive values for
56:
Ending Spam: Bayesian
Content Filtering and the Art of Statistical Language Classification
94:
47:
27:
137:
90:
31:
35:
26:
SBPH is a way of generating a large number of features from an incoming
50:(some article text comes from this document, which is under the GFDL)
23:
that can match mutating phrases as well as single words.
98:
118:
8:
48:A paper on the subject as it relates to spam
125:
111:
59:. No Starch Press. 2005. p. 108.
7:
79:
77:
14:
81:
17:Sparse binary polynomial hashing
30:automatically, and then using
19:(SBPH) is a generalization of
1:
97:. You can help Knowledge by
170:
76:
21:Bayesian spam filtering
93:-related article is a
38:/nonspam evaluation.
144:Bayesian statistics
106:
105:
66:978-1-59327-052-0
161:
154:Statistics stubs
127:
120:
113:
85:
78:
70:
169:
168:
164:
163:
162:
160:
159:
158:
134:
133:
132:
131:
74:
67:
53:
44:
12:
11:
5:
167:
165:
157:
156:
151:
146:
136:
135:
130:
129:
122:
115:
107:
104:
103:
86:
72:
71:
65:
51:
43:
42:External links
40:
13:
10:
9:
6:
4:
3:
2:
166:
155:
152:
150:
147:
145:
142:
141:
139:
128:
123:
121:
116:
114:
109:
108:
102:
100:
96:
92:
87:
84:
80:
75:
68:
62:
58:
57:
52:
49:
46:
45:
41:
39:
37:
33:
29:
24:
22:
18:
99:expanding it
88:
73:
55:
25:
16:
15:
138:Categories
91:statistics
32:statistics
149:Anti-spam
63:
89:This
95:stub
61:ISBN
36:spam
28:text
140::
126:e
119:t
112:v
101:.
69:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.