classifier 1.3.0 → 1.3.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/README +2 -1
- data/Rakefile +3 -4
- data/bin/bayes.rb +1 -1
- data/bin/summarize.rb +7 -2
- data/lib/classifier/extensions/string.rb +2 -2
- data/lib/classifier/extensions/vector_serialize.rb +2 -2
- data/lib/classifier/lsi.rb +2 -2
- data/lib/classifier/lsi/content_node.rb +1 -1
- metadata +73 -159
- data/doc/classes/Array.html +0 -139
- data/doc/classes/Array.src/M000003.html +0 -18
- data/doc/classes/Classifier.html +0 -140
- data/doc/classes/Classifier/Bayes.html +0 -317
- data/doc/classes/Classifier/Bayes.src/M000038.html +0 -20
- data/doc/classes/Classifier/Bayes.src/M000039.html +0 -23
- data/doc/classes/Classifier/Bayes.src/M000040.html +0 -30
- data/doc/classes/Classifier/Bayes.src/M000041.html +0 -27
- data/doc/classes/Classifier/Bayes.src/M000042.html +0 -18
- data/doc/classes/Classifier/Bayes.src/M000043.html +0 -25
- data/doc/classes/Classifier/Bayes.src/M000044.html +0 -18
- data/doc/classes/Classifier/ContentNode.html +0 -247
- data/doc/classes/Classifier/ContentNode.src/M000046.html +0 -19
- data/doc/classes/Classifier/ContentNode.src/M000047.html +0 -18
- data/doc/classes/Classifier/ContentNode.src/M000048.html +0 -18
- data/doc/classes/Classifier/ContentNode.src/M000049.html +0 -49
- data/doc/classes/Classifier/LSI.html +0 -539
- data/doc/classes/Classifier/LSI.src/M000022.html +0 -20
- data/doc/classes/Classifier/LSI.src/M000023.html +0 -18
- data/doc/classes/Classifier/LSI.src/M000024.html +0 -21
- data/doc/classes/Classifier/LSI.src/M000025.html +0 -18
- data/doc/classes/Classifier/LSI.src/M000026.html +0 -19
- data/doc/classes/Classifier/LSI.src/M000027.html +0 -21
- data/doc/classes/Classifier/LSI.src/M000028.html +0 -18
- data/doc/classes/Classifier/LSI.src/M000029.html +0 -19
- data/doc/classes/Classifier/LSI.src/M000030.html +0 -43
- data/doc/classes/Classifier/LSI.src/M000031.html +0 -23
- data/doc/classes/Classifier/LSI.src/M000032.html +0 -30
- data/doc/classes/Classifier/LSI.src/M000033.html +0 -30
- data/doc/classes/Classifier/LSI.src/M000034.html +0 -21
- data/doc/classes/Classifier/LSI.src/M000035.html +0 -21
- data/doc/classes/Classifier/LSI.src/M000036.html +0 -31
- data/doc/classes/Classifier/LSI.src/M000037.html +0 -21
- data/doc/classes/Classifier/WordList.html +0 -217
- data/doc/classes/Classifier/WordList.src/M000017.html +0 -18
- data/doc/classes/Classifier/WordList.src/M000018.html +0 -19
- data/doc/classes/Classifier/WordList.src/M000019.html +0 -19
- data/doc/classes/Classifier/WordList.src/M000020.html +0 -18
- data/doc/classes/Classifier/WordList.src/M000021.html +0 -18
- data/doc/classes/GSL.html +0 -112
- data/doc/classes/GSL/Matrix.html +0 -126
- data/doc/classes/GSL/Vector.html +0 -156
- data/doc/classes/GSL/Vector.src/M000015.html +0 -18
- data/doc/classes/GSL/Vector.src/M000016.html +0 -19
- data/doc/classes/Matrix.html +0 -184
- data/doc/classes/Matrix.src/M000004.html +0 -18
- data/doc/classes/Matrix.src/M000005.html +0 -76
- data/doc/classes/Matrix.src/M000006.html +0 -18
- data/doc/classes/Object.html +0 -139
- data/doc/classes/Object.src/M000007.html +0 -16
- data/doc/classes/String.html +0 -275
- data/doc/classes/String.src/M000008.html +0 -18
- data/doc/classes/String.src/M000009.html +0 -18
- data/doc/classes/String.src/M000010.html +0 -18
- data/doc/classes/String.src/M000011.html +0 -18
- data/doc/classes/String.src/M000012.html +0 -18
- data/doc/classes/String.src/M000013.html +0 -18
- data/doc/classes/String.src/M000014.html +0 -18
- data/doc/classes/Vector.html +0 -154
- data/doc/classes/Vector.src/M000001.html +0 -22
- data/doc/classes/Vector.src/M000002.html +0 -25
- data/doc/created.rid +0 -1
- data/doc/files/README.html +0 -252
- data/doc/files/lib/classifier/bayes_rb.html +0 -115
- data/doc/files/lib/classifier/extensions/string_rb.html +0 -122
- data/doc/files/lib/classifier/extensions/vector_rb.html +0 -120
- data/doc/files/lib/classifier/extensions/vector_serialize_rb.html +0 -101
- data/doc/files/lib/classifier/extensions/word_hash_rb.html +0 -115
- data/doc/files/lib/classifier/lsi/content_node_rb.html +0 -115
- data/doc/files/lib/classifier/lsi/summary_rb.html +0 -115
- data/doc/files/lib/classifier/lsi/word_list_rb.html +0 -115
- data/doc/files/lib/classifier/lsi_rb.html +0 -127
- data/doc/files/lib/classifier_rb.html +0 -125
- data/doc/fr_class_index.html +0 -39
- data/doc/fr_file_index.html +0 -37
- data/doc/fr_method_index.html +0 -75
- data/doc/index.html +0 -24
- data/doc/rdoc-style.css +0 -208
|
@@ -1,16 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>prepare_category_name (Object)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/extensions/string.rb, line 15</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">prepare_category_name</span>; <span class="ruby-identifier">to_s</span>.<span class="ruby-identifier">gsub</span>(<span class="ruby-value str">"_"</span>,<span class="ruby-value str">" "</span>).<span class="ruby-identifier">capitalize</span>.<span class="ruby-identifier">intern</span> <span class="ruby-keyword kw">end</span></pre>
|
|
15
|
-
</body>
|
|
16
|
-
</html>
|
data/doc/classes/String.html
DELETED
|
@@ -1,275 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
|
|
7
|
-
<head>
|
|
8
|
-
<title>Class: String</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<meta http-equiv="Content-Script-Type" content="text/javascript" />
|
|
11
|
-
<link rel="stylesheet" href=".././rdoc-style.css" type="text/css" media="screen" />
|
|
12
|
-
<script type="text/javascript">
|
|
13
|
-
// <![CDATA[
|
|
14
|
-
|
|
15
|
-
function popupCode( url ) {
|
|
16
|
-
window.open(url, "Code", "resizable=yes,scrollbars=yes,toolbar=no,status=no,height=150,width=400")
|
|
17
|
-
}
|
|
18
|
-
|
|
19
|
-
function toggleCode( id ) {
|
|
20
|
-
if ( document.getElementById )
|
|
21
|
-
elem = document.getElementById( id );
|
|
22
|
-
else if ( document.all )
|
|
23
|
-
elem = eval( "document.all." + id );
|
|
24
|
-
else
|
|
25
|
-
return false;
|
|
26
|
-
|
|
27
|
-
elemStyle = elem.style;
|
|
28
|
-
|
|
29
|
-
if ( elemStyle.display != "block" ) {
|
|
30
|
-
elemStyle.display = "block"
|
|
31
|
-
} else {
|
|
32
|
-
elemStyle.display = "none"
|
|
33
|
-
}
|
|
34
|
-
|
|
35
|
-
return true;
|
|
36
|
-
}
|
|
37
|
-
|
|
38
|
-
// Make codeblocks hidden by default
|
|
39
|
-
document.writeln( "<style type=\"text/css\">div.method-source-code { display: none }</style>" )
|
|
40
|
-
|
|
41
|
-
// ]]>
|
|
42
|
-
</script>
|
|
43
|
-
|
|
44
|
-
</head>
|
|
45
|
-
<body>
|
|
46
|
-
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
<div id="classHeader">
|
|
50
|
-
<table class="header-table">
|
|
51
|
-
<tr class="top-aligned-row">
|
|
52
|
-
<td><strong>Class</strong></td>
|
|
53
|
-
<td class="class-name-in-header">String</td>
|
|
54
|
-
</tr>
|
|
55
|
-
<tr class="top-aligned-row">
|
|
56
|
-
<td><strong>In:</strong></td>
|
|
57
|
-
<td>
|
|
58
|
-
<a href="../files/lib/classifier/extensions/word_hash_rb.html">
|
|
59
|
-
lib/classifier/extensions/word_hash.rb
|
|
60
|
-
</a>
|
|
61
|
-
<br />
|
|
62
|
-
<a href="../files/lib/classifier/lsi/summary_rb.html">
|
|
63
|
-
lib/classifier/lsi/summary.rb
|
|
64
|
-
</a>
|
|
65
|
-
<br />
|
|
66
|
-
</td>
|
|
67
|
-
</tr>
|
|
68
|
-
|
|
69
|
-
<tr class="top-aligned-row">
|
|
70
|
-
<td><strong>Parent:</strong></td>
|
|
71
|
-
<td>
|
|
72
|
-
<a href="Object.html">
|
|
73
|
-
Object
|
|
74
|
-
</a>
|
|
75
|
-
</td>
|
|
76
|
-
</tr>
|
|
77
|
-
</table>
|
|
78
|
-
</div>
|
|
79
|
-
<!-- banner header -->
|
|
80
|
-
|
|
81
|
-
<div id="bodyContent">
|
|
82
|
-
|
|
83
|
-
|
|
84
|
-
|
|
85
|
-
<div id="contextContent">
|
|
86
|
-
|
|
87
|
-
<div id="description">
|
|
88
|
-
<table>
|
|
89
|
-
<tr><td valign="top">Author:</td><td>Lucas Carlson (<a href="mailto:lucas@rufy.com">lucas@rufy.com</a>)
|
|
90
|
-
|
|
91
|
-
</td></tr>
|
|
92
|
-
<tr><td valign="top">Copyright:</td><td>Copyright © 2005 Lucas Carlson
|
|
93
|
-
|
|
94
|
-
</td></tr>
|
|
95
|
-
<tr><td valign="top">License:</td><td>LGPL
|
|
96
|
-
|
|
97
|
-
</td></tr>
|
|
98
|
-
</table>
|
|
99
|
-
|
|
100
|
-
</div>
|
|
101
|
-
|
|
102
|
-
|
|
103
|
-
</div>
|
|
104
|
-
|
|
105
|
-
<div id="method-list">
|
|
106
|
-
<h3 class="section-bar">Methods</h3>
|
|
107
|
-
|
|
108
|
-
<div class="name-list">
|
|
109
|
-
<a href="#M000010">clean_word_hash</a>
|
|
110
|
-
<a href="#M000012">paragraph_summary</a>
|
|
111
|
-
<a href="#M000014">split_paragraphs</a>
|
|
112
|
-
<a href="#M000013">split_sentences</a>
|
|
113
|
-
<a href="#M000011">summary</a>
|
|
114
|
-
<a href="#M000008">without_punctuation</a>
|
|
115
|
-
<a href="#M000009">word_hash</a>
|
|
116
|
-
</div>
|
|
117
|
-
</div>
|
|
118
|
-
|
|
119
|
-
</div>
|
|
120
|
-
|
|
121
|
-
|
|
122
|
-
<!-- if includes -->
|
|
123
|
-
|
|
124
|
-
<div id="section">
|
|
125
|
-
|
|
126
|
-
|
|
127
|
-
<div id="constants-list">
|
|
128
|
-
<h3 class="section-bar">Constants</h3>
|
|
129
|
-
|
|
130
|
-
<div class="name-list">
|
|
131
|
-
<table summary="Constants">
|
|
132
|
-
<tr class="top-aligned-row context-row">
|
|
133
|
-
<td class="context-item-name">CORPUS_SKIP_WORDS</td>
|
|
134
|
-
<td>=</td>
|
|
135
|
-
<td class="context-item-value">[ "a", "again", "all", "along", "are", "also", "an", "and", "as", "at", "but", "by", "came", "can", "cant", "couldnt", "did", "didn", "didnt", "do", "doesnt", "dont", "ever", "first", "from", "have", "her", "here", "him", "how", "i", "if", "in", "into", "is", "isnt", "it", "itll", "just", "last", "least", "like", "most", "my", "new", "no", "not", "now", "of", "on", "or", "should", "sinc", "so", "some", "th", "than", "this", "that", "the", "their", "then", "those", "to", "told", "too", "true", "try", "until", "url", "us", "were", "when", "whether", "while", "with", "within", "yes", "you", "youll", ]</td>
|
|
136
|
-
</tr>
|
|
137
|
-
</table>
|
|
138
|
-
</div>
|
|
139
|
-
</div>
|
|
140
|
-
|
|
141
|
-
|
|
142
|
-
|
|
143
|
-
|
|
144
|
-
|
|
145
|
-
|
|
146
|
-
<!-- if method_list -->
|
|
147
|
-
<div id="methods">
|
|
148
|
-
<h3 class="section-bar">Public Instance methods</h3>
|
|
149
|
-
|
|
150
|
-
<div id="method-M000010" class="method-detail">
|
|
151
|
-
<a name="M000010"></a>
|
|
152
|
-
|
|
153
|
-
<div class="method-heading">
|
|
154
|
-
<a href="String.src/M000010.html" target="Code" class="method-signature"
|
|
155
|
-
onclick="popupCode('String.src/M000010.html');return false;">
|
|
156
|
-
<span class="method-name">clean_word_hash</span><span class="method-args">()</span>
|
|
157
|
-
</a>
|
|
158
|
-
</div>
|
|
159
|
-
|
|
160
|
-
<div class="method-description">
|
|
161
|
-
<p>
|
|
162
|
-
Return a word hash without extra punctuation or short symbols, just stemmed
|
|
163
|
-
words
|
|
164
|
-
</p>
|
|
165
|
-
</div>
|
|
166
|
-
</div>
|
|
167
|
-
|
|
168
|
-
<div id="method-M000012" class="method-detail">
|
|
169
|
-
<a name="M000012"></a>
|
|
170
|
-
|
|
171
|
-
<div class="method-heading">
|
|
172
|
-
<a href="String.src/M000012.html" target="Code" class="method-signature"
|
|
173
|
-
onclick="popupCode('String.src/M000012.html');return false;">
|
|
174
|
-
<span class="method-name">paragraph_summary</span><span class="method-args">( count=1, separator=" [...] " )</span>
|
|
175
|
-
</a>
|
|
176
|
-
</div>
|
|
177
|
-
|
|
178
|
-
<div class="method-description">
|
|
179
|
-
</div>
|
|
180
|
-
</div>
|
|
181
|
-
|
|
182
|
-
<div id="method-M000014" class="method-detail">
|
|
183
|
-
<a name="M000014"></a>
|
|
184
|
-
|
|
185
|
-
<div class="method-heading">
|
|
186
|
-
<a href="String.src/M000014.html" target="Code" class="method-signature"
|
|
187
|
-
onclick="popupCode('String.src/M000014.html');return false;">
|
|
188
|
-
<span class="method-name">split_paragraphs</span><span class="method-args">()</span>
|
|
189
|
-
</a>
|
|
190
|
-
</div>
|
|
191
|
-
|
|
192
|
-
<div class="method-description">
|
|
193
|
-
</div>
|
|
194
|
-
</div>
|
|
195
|
-
|
|
196
|
-
<div id="method-M000013" class="method-detail">
|
|
197
|
-
<a name="M000013"></a>
|
|
198
|
-
|
|
199
|
-
<div class="method-heading">
|
|
200
|
-
<a href="String.src/M000013.html" target="Code" class="method-signature"
|
|
201
|
-
onclick="popupCode('String.src/M000013.html');return false;">
|
|
202
|
-
<span class="method-name">split_sentences</span><span class="method-args">()</span>
|
|
203
|
-
</a>
|
|
204
|
-
</div>
|
|
205
|
-
|
|
206
|
-
<div class="method-description">
|
|
207
|
-
</div>
|
|
208
|
-
</div>
|
|
209
|
-
|
|
210
|
-
<div id="method-M000011" class="method-detail">
|
|
211
|
-
<a name="M000011"></a>
|
|
212
|
-
|
|
213
|
-
<div class="method-heading">
|
|
214
|
-
<a href="String.src/M000011.html" target="Code" class="method-signature"
|
|
215
|
-
onclick="popupCode('String.src/M000011.html');return false;">
|
|
216
|
-
<span class="method-name">summary</span><span class="method-args">( count=10, separator=" [...] " )</span>
|
|
217
|
-
</a>
|
|
218
|
-
</div>
|
|
219
|
-
|
|
220
|
-
<div class="method-description">
|
|
221
|
-
</div>
|
|
222
|
-
</div>
|
|
223
|
-
|
|
224
|
-
<div id="method-M000008" class="method-detail">
|
|
225
|
-
<a name="M000008"></a>
|
|
226
|
-
|
|
227
|
-
<div class="method-heading">
|
|
228
|
-
<a href="String.src/M000008.html" target="Code" class="method-signature"
|
|
229
|
-
onclick="popupCode('String.src/M000008.html');return false;">
|
|
230
|
-
<span class="method-name">without_punctuation</span><span class="method-args">()</span>
|
|
231
|
-
</a>
|
|
232
|
-
</div>
|
|
233
|
-
|
|
234
|
-
<div class="method-description">
|
|
235
|
-
<p>
|
|
236
|
-
Removes common punctuation symbols, returning a new string. E.g.,
|
|
237
|
-
</p>
|
|
238
|
-
<pre>
|
|
239
|
-
"Hello (greeting's), with {braces} < >...?".without_punctuation
|
|
240
|
-
=> "Hello greetings with braces "
|
|
241
|
-
</pre>
|
|
242
|
-
</div>
|
|
243
|
-
</div>
|
|
244
|
-
|
|
245
|
-
<div id="method-M000009" class="method-detail">
|
|
246
|
-
<a name="M000009"></a>
|
|
247
|
-
|
|
248
|
-
<div class="method-heading">
|
|
249
|
-
<a href="String.src/M000009.html" target="Code" class="method-signature"
|
|
250
|
-
onclick="popupCode('String.src/M000009.html');return false;">
|
|
251
|
-
<span class="method-name">word_hash</span><span class="method-args">()</span>
|
|
252
|
-
</a>
|
|
253
|
-
</div>
|
|
254
|
-
|
|
255
|
-
<div class="method-description">
|
|
256
|
-
<p>
|
|
257
|
-
Return a Hash of strings => ints. Each word in the string is stemmed,
|
|
258
|
-
interned, and indexes to its frequency in the document.
|
|
259
|
-
</p>
|
|
260
|
-
</div>
|
|
261
|
-
</div>
|
|
262
|
-
|
|
263
|
-
|
|
264
|
-
</div>
|
|
265
|
-
|
|
266
|
-
|
|
267
|
-
</div>
|
|
268
|
-
|
|
269
|
-
|
|
270
|
-
<div id="validator-badges">
|
|
271
|
-
<p><small><a href="http://validator.w3.org/check/referer">[Validate]</a></small></p>
|
|
272
|
-
</div>
|
|
273
|
-
|
|
274
|
-
</body>
|
|
275
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>without_punctuation (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/extensions/word_hash.rb, line 13</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">without_punctuation</span>
|
|
15
|
-
<span class="ruby-identifier">tr</span>( <span class="ruby-value str">',?.!;:"@#$%^&*()_=+[]{}\|<>/`~'</span>, <span class="ruby-value str">" "</span> ) .<span class="ruby-identifier">tr</span>( <span class="ruby-value str">"'\-"</span>, <span class="ruby-value str">""</span>)
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>word_hash (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/extensions/word_hash.rb, line 19</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">word_hash</span>
|
|
15
|
-
<span class="ruby-identifier">word_hash_for_words</span>(<span class="ruby-identifier">gsub</span>(<span class="ruby-regexp re">/[^\w\s]/</span>,<span class="ruby-value str">""</span>).<span class="ruby-identifier">split</span> <span class="ruby-operator">+</span> <span class="ruby-identifier">gsub</span>(<span class="ruby-regexp re">/[\w]/</span>,<span class="ruby-value str">" "</span>).<span class="ruby-identifier">split</span>)
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>clean_word_hash (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/extensions/word_hash.rb, line 24</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">clean_word_hash</span>
|
|
15
|
-
<span class="ruby-identifier">word_hash_for_words</span> <span class="ruby-identifier">gsub</span>(<span class="ruby-regexp re">/[^\w\s]/</span>,<span class="ruby-value str">""</span>).<span class="ruby-identifier">split</span>
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>summary (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/lsi/summary.rb, line 6</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">summary</span>( <span class="ruby-identifier">count</span>=<span class="ruby-value">10</span>, <span class="ruby-identifier">separator</span>=<span class="ruby-value str">" [...] "</span> )
|
|
15
|
-
<span class="ruby-identifier">perform_lsi</span> <span class="ruby-identifier">split_sentences</span>, <span class="ruby-identifier">count</span>, <span class="ruby-identifier">separator</span>
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>paragraph_summary (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/lsi/summary.rb, line 10</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">paragraph_summary</span>( <span class="ruby-identifier">count</span>=<span class="ruby-value">1</span>, <span class="ruby-identifier">separator</span>=<span class="ruby-value str">" [...] "</span> )
|
|
15
|
-
<span class="ruby-identifier">perform_lsi</span> <span class="ruby-identifier">split_paragraphs</span>, <span class="ruby-identifier">count</span>, <span class="ruby-identifier">separator</span>
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>split_sentences (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/lsi/summary.rb, line 14</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">split_sentences</span>
|
|
15
|
-
<span class="ruby-identifier">split</span> <span class="ruby-regexp re">/(\.|\!|\?)/</span> <span class="ruby-comment cmt"># TODO: make this less primitive</span>
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
|
@@ -1,18 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html>
|
|
7
|
-
<head>
|
|
8
|
-
<title>split_paragraphs (String)</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<link rel="stylesheet" href="../.././rdoc-style.css" type="text/css" media="screen" />
|
|
11
|
-
</head>
|
|
12
|
-
<body class="standalone-code">
|
|
13
|
-
<pre><span class="ruby-comment cmt"># File lib/classifier/lsi/summary.rb, line 18</span>
|
|
14
|
-
<span class="ruby-keyword kw">def</span> <span class="ruby-identifier">split_paragraphs</span>
|
|
15
|
-
<span class="ruby-identifier">split</span> <span class="ruby-regexp re">/(\n\n|\r\r|\r\n\r\n)/</span> <span class="ruby-comment cmt"># TODO: make this less primitive</span>
|
|
16
|
-
<span class="ruby-keyword kw">end</span></pre>
|
|
17
|
-
</body>
|
|
18
|
-
</html>
|
data/doc/classes/Vector.html
DELETED
|
@@ -1,154 +0,0 @@
|
|
|
1
|
-
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
2
|
-
<!DOCTYPE html
|
|
3
|
-
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
|
|
4
|
-
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
5
|
-
|
|
6
|
-
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
|
|
7
|
-
<head>
|
|
8
|
-
<title>Class: Vector</title>
|
|
9
|
-
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
10
|
-
<meta http-equiv="Content-Script-Type" content="text/javascript" />
|
|
11
|
-
<link rel="stylesheet" href=".././rdoc-style.css" type="text/css" media="screen" />
|
|
12
|
-
<script type="text/javascript">
|
|
13
|
-
// <![CDATA[
|
|
14
|
-
|
|
15
|
-
function popupCode( url ) {
|
|
16
|
-
window.open(url, "Code", "resizable=yes,scrollbars=yes,toolbar=no,status=no,height=150,width=400")
|
|
17
|
-
}
|
|
18
|
-
|
|
19
|
-
function toggleCode( id ) {
|
|
20
|
-
if ( document.getElementById )
|
|
21
|
-
elem = document.getElementById( id );
|
|
22
|
-
else if ( document.all )
|
|
23
|
-
elem = eval( "document.all." + id );
|
|
24
|
-
else
|
|
25
|
-
return false;
|
|
26
|
-
|
|
27
|
-
elemStyle = elem.style;
|
|
28
|
-
|
|
29
|
-
if ( elemStyle.display != "block" ) {
|
|
30
|
-
elemStyle.display = "block"
|
|
31
|
-
} else {
|
|
32
|
-
elemStyle.display = "none"
|
|
33
|
-
}
|
|
34
|
-
|
|
35
|
-
return true;
|
|
36
|
-
}
|
|
37
|
-
|
|
38
|
-
// Make codeblocks hidden by default
|
|
39
|
-
document.writeln( "<style type=\"text/css\">div.method-source-code { display: none }</style>" )
|
|
40
|
-
|
|
41
|
-
// ]]>
|
|
42
|
-
</script>
|
|
43
|
-
|
|
44
|
-
</head>
|
|
45
|
-
<body>
|
|
46
|
-
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
<div id="classHeader">
|
|
50
|
-
<table class="header-table">
|
|
51
|
-
<tr class="top-aligned-row">
|
|
52
|
-
<td><strong>Class</strong></td>
|
|
53
|
-
<td class="class-name-in-header">Vector</td>
|
|
54
|
-
</tr>
|
|
55
|
-
<tr class="top-aligned-row">
|
|
56
|
-
<td><strong>In:</strong></td>
|
|
57
|
-
<td>
|
|
58
|
-
<a href="../files/lib/classifier/extensions/vector_rb.html">
|
|
59
|
-
lib/classifier/extensions/vector.rb
|
|
60
|
-
</a>
|
|
61
|
-
<br />
|
|
62
|
-
</td>
|
|
63
|
-
</tr>
|
|
64
|
-
|
|
65
|
-
<tr class="top-aligned-row">
|
|
66
|
-
<td><strong>Parent:</strong></td>
|
|
67
|
-
<td>
|
|
68
|
-
<a href="Object.html">
|
|
69
|
-
Object
|
|
70
|
-
</a>
|
|
71
|
-
</td>
|
|
72
|
-
</tr>
|
|
73
|
-
</table>
|
|
74
|
-
</div>
|
|
75
|
-
<!-- banner header -->
|
|
76
|
-
|
|
77
|
-
<div id="bodyContent">
|
|
78
|
-
|
|
79
|
-
|
|
80
|
-
|
|
81
|
-
<div id="contextContent">
|
|
82
|
-
|
|
83
|
-
|
|
84
|
-
|
|
85
|
-
</div>
|
|
86
|
-
|
|
87
|
-
<div id="method-list">
|
|
88
|
-
<h3 class="section-bar">Methods</h3>
|
|
89
|
-
|
|
90
|
-
<div class="name-list">
|
|
91
|
-
<a href="#M000001">magnitude</a>
|
|
92
|
-
<a href="#M000002">normalize</a>
|
|
93
|
-
</div>
|
|
94
|
-
</div>
|
|
95
|
-
|
|
96
|
-
</div>
|
|
97
|
-
|
|
98
|
-
|
|
99
|
-
<!-- if includes -->
|
|
100
|
-
|
|
101
|
-
<div id="section">
|
|
102
|
-
|
|
103
|
-
|
|
104
|
-
|
|
105
|
-
|
|
106
|
-
|
|
107
|
-
|
|
108
|
-
|
|
109
|
-
|
|
110
|
-
<!-- if method_list -->
|
|
111
|
-
<div id="methods">
|
|
112
|
-
<h3 class="section-bar">Public Instance methods</h3>
|
|
113
|
-
|
|
114
|
-
<div id="method-M000001" class="method-detail">
|
|
115
|
-
<a name="M000001"></a>
|
|
116
|
-
|
|
117
|
-
<div class="method-heading">
|
|
118
|
-
<a href="Vector.src/M000001.html" target="Code" class="method-signature"
|
|
119
|
-
onclick="popupCode('Vector.src/M000001.html');return false;">
|
|
120
|
-
<span class="method-name">magnitude</span><span class="method-args">()</span>
|
|
121
|
-
</a>
|
|
122
|
-
</div>
|
|
123
|
-
|
|
124
|
-
<div class="method-description">
|
|
125
|
-
</div>
|
|
126
|
-
</div>
|
|
127
|
-
|
|
128
|
-
<div id="method-M000002" class="method-detail">
|
|
129
|
-
<a name="M000002"></a>
|
|
130
|
-
|
|
131
|
-
<div class="method-heading">
|
|
132
|
-
<a href="Vector.src/M000002.html" target="Code" class="method-signature"
|
|
133
|
-
onclick="popupCode('Vector.src/M000002.html');return false;">
|
|
134
|
-
<span class="method-name">normalize</span><span class="method-args">()</span>
|
|
135
|
-
</a>
|
|
136
|
-
</div>
|
|
137
|
-
|
|
138
|
-
<div class="method-description">
|
|
139
|
-
</div>
|
|
140
|
-
</div>
|
|
141
|
-
|
|
142
|
-
|
|
143
|
-
</div>
|
|
144
|
-
|
|
145
|
-
|
|
146
|
-
</div>
|
|
147
|
-
|
|
148
|
-
|
|
149
|
-
<div id="validator-badges">
|
|
150
|
-
<p><small><a href="http://validator.w3.org/check/referer">[Validate]</a></small></p>
|
|
151
|
-
</div>
|
|
152
|
-
|
|
153
|
-
</body>
|
|
154
|
-
</html>
|