-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy pathtwenty_newsgroups.html
428 lines (387 loc) · 31.3 KB
/
twenty_newsgroups.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>The 20 newsgroups text dataset — scikit-learn 0.15-git documentation</title>
<!-- htmltitle is before nature.css - we use this hack to load bootstrap first -->
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<link rel="stylesheet" href="../_static/css/bootstrap.min.css" media="screen" />
<link rel="stylesheet" href="../_static/css/bootstrap-responsive.css"/>
<link rel="stylesheet" href="../_static/nature.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: '../',
VERSION: '0.15-git',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true
};
</script>
<script type="text/javascript" src="../_static/jquery.js"></script>
<script type="text/javascript" src="../_static/underscore.js"></script>
<script type="text/javascript" src="../_static/doctools.js"></script>
<script type="text/javascript" src="../_static/js/copybutton.js"></script>
<link rel="shortcut icon" href="../_static/favicon.ico"/>
<link rel="author" title="About these documents" href="../about.html" />
<link rel="top" title="scikit-learn 0.15-git documentation" href="../index.html" />
<link rel="next" title="Reference" href="../modules/classes.html" />
<link rel="prev" title="The Olivetti faces dataset" href="olivetti_faces.html" />
<script type="text/javascript" src="../_static/sidebar.js"></script>
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<script src="../_static/js/bootstrap.min.js" type="text/javascript"></script>
<link rel="canonical" href="https://scikit-learn.org/stable/datasets/twenty_newsgroups.html" />
<script type="text/javascript">
$("div.buttonNext, div.buttonPrevious").hover(
function () {
$(this).css('background-color', '#FF9C34');
},
function () {
$(this).css('background-color', '#A7D6E2');
}
);
var bodywrapper = $('.bodywrapper');
var sidebarbutton = $('#sidebarbutton');
sidebarbutton.css({'height': '900px'});
</script>
</head>
<body>
<div class="header-wrapper">
<div class="header">
<p class="logo"><a href="../index.html">
<img src="../_static/scikit-learn-logo-small.png" alt="Logo"/>
</a>
</p><div class="navbar">
<ul>
<li><a href="../../stable/index.html">Home</a></li>
<li><a href="../../stable/install.html">Installation</a></li>
<li class="btn-li"><div class="btn-group">
<a href="../documentation.html">Documentation</a>
<a class="btn dropdown-toggle" data-toggle="dropdown">
<span class="caret"></span>
</a>
<ul class="dropdown-menu">
<li class="link-title">Scikit-learn 0.15 (stable)</li>
<li><a href="../tutorial/index.html">Tutorials</a></li>
<li><a href="../user_guide.html">User guide</a></li>
<li><a href="../modules/classes.html">API</a></li>
<li><a href="../faq.html">FAQ</a></li>
<li class="divider"></li>
<li><a href="http://scikit-learn.org/dev/documentation.html">Development</a></li>
</ul>
</div>
</li>
<li><a href="../auto_examples/index.html">Examples</a></li>
</ul>
<div class="search_form">
<div id="cse" style="width: 100%;"></div>
</div>
</div> <!-- end navbar --></div>
</div>
<!-- Github "fork me" ribbon -->
<a href="https://github.com/scikit-learn/scikit-learn">
<img class="fork-me"
style="position: absolute; top: 0; right: 0; border: 0;"
src="../_static/img/forkme.png"
alt="Fork me on GitHub" />
</a>
<div class="content-wrapper">
<div class="sphinxsidebar">
<div class="sphinxsidebarwrapper">
<div class="rel rellarge">
<!-- rellinks[1:] is an ugly hack to avoid link to module
index -->
<div class="rellink">
<a href="olivetti_faces.html"
accesskey="P">Previous
<br/>
<span class="smallrellink">
The Olivetti fac...
</span>
<span class="hiddenrellink">
The Olivetti faces dataset
</span>
</a>
</div>
<div class="spacer">
</div>
<div class="rellink">
<a href="../modules/classes.html"
accesskey="N">Next
<br/>
<span class="smallrellink">
Reference
</span>
<span class="hiddenrellink">
Reference
</span>
</a>
</div>
<!-- Ad a link to the 'up' page -->
</div>
<p class="doc-version">This documentation is for scikit-learn <strong>version 0.15-git</strong> — <a href="http://scikit-learn.org/stable/support.html#documentation-resources">Other versions</a></p>
<p class="citing">If you use the software, please consider <a href="../about.html#citing-scikit-learn">citing scikit-learn</a>.</p>
<ul>
<li><a class="reference internal" href="#">The 20 newsgroups text dataset</a><ul>
<li><a class="reference internal" href="#usage">Usage</a></li>
<li><a class="reference internal" href="#converting-text-to-vectors">Converting text to vectors</a></li>
<li><a class="reference internal" href="#filtering-text-for-more-realistic-training">Filtering text for more realistic training</a></li>
</ul>
</li>
</ul>
</div>
</div>
<div class="content">
<div class="documentwrapper">
<div class="bodywrapper">
<div class="body">
<div class="section" id="the-20-newsgroups-text-dataset">
<span id="newsgroups"></span><h1>The 20 newsgroups text dataset<a class="headerlink" href="#the-20-newsgroups-text-dataset" title="Permalink to this headline">¶</a></h1>
<p>The 20 newsgroups dataset comprises around 18000 newsgroups posts on
20 topics split in two subsets: one for training (or development)
and the other one for testing (or for performance evaluation). The split
between the train and test set is based upon a messages posted before
and after a specific date.</p>
<p>This module contains two loaders. The first one,
<tt class="docutils literal"><span class="pre">sklearn.datasets.fetch_20newsgroups</span></tt>,
returns a list of the raw texts that can be fed to text feature
extractors such as <tt class="xref py py-class docutils literal"><span class="pre">sklearn.feature_extraction.text.Vectorizer</span></tt>
with custom parameters so as to extract feature vectors.
The second one, <tt class="docutils literal"><span class="pre">sklearn.datasets.fetch_20newsgroups_vectorized</span></tt>,
returns ready-to-use features, i.e., it is not necessary to use a feature
extractor.</p>
<div class="section" id="usage">
<h2>Usage<a class="headerlink" href="#usage" title="Permalink to this headline">¶</a></h2>
<p>The <tt class="docutils literal"><span class="pre">sklearn.datasets.fetch_20newsgroups</span></tt> function is a data
fetching / caching functions that downloads the data archive from
the original <a class="reference external" href="http://people.csail.mit.edu/jrennie/20Newsgroups/">20 newsgroups website</a>, extracts the archive contents
in the <tt class="docutils literal"><span class="pre">~/scikit_learn_data/20news_home</span></tt> folder and calls the
<tt class="docutils literal"><span class="pre">sklearn.datasets.load_file</span></tt> on either the training or
testing set folder, or both of them:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <span class="n">fetch_20newsgroups</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span> <span class="o">=</span> <span class="n">fetch_20newsgroups</span><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s">'train'</span><span class="p">)</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
<span class="gp">>>> </span><span class="n">pprint</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target_names</span><span class="p">))</span>
<span class="go">['alt.atheism',</span>
<span class="go"> 'comp.graphics',</span>
<span class="go"> 'comp.os.ms-windows.misc',</span>
<span class="go"> 'comp.sys.ibm.pc.hardware',</span>
<span class="go"> 'comp.sys.mac.hardware',</span>
<span class="go"> 'comp.windows.x',</span>
<span class="go"> 'misc.forsale',</span>
<span class="go"> 'rec.autos',</span>
<span class="go"> 'rec.motorcycles',</span>
<span class="go"> 'rec.sport.baseball',</span>
<span class="go"> 'rec.sport.hockey',</span>
<span class="go"> 'sci.crypt',</span>
<span class="go"> 'sci.electronics',</span>
<span class="go"> 'sci.med',</span>
<span class="go"> 'sci.space',</span>
<span class="go"> 'soc.religion.christian',</span>
<span class="go"> 'talk.politics.guns',</span>
<span class="go"> 'talk.politics.mideast',</span>
<span class="go"> 'talk.politics.misc',</span>
<span class="go"> 'talk.religion.misc']</span>
</pre></div>
</div>
<p>The real data lies in the <tt class="docutils literal"><span class="pre">filenames</span></tt> and <tt class="docutils literal"><span class="pre">target</span></tt> attributes. The target
attribute is the integer index of the category:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">filenames</span><span class="o">.</span><span class="n">shape</span>
<span class="go">(11314,)</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target</span><span class="o">.</span><span class="n">shape</span>
<span class="go">(11314,)</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target</span><span class="p">[:</span><span class="mi">10</span><span class="p">]</span>
<span class="go">array([12, 6, 9, 8, 6, 7, 9, 2, 13, 19])</span>
</pre></div>
</div>
<p>It is possible to load only a sub-selection of the categories by passing the
list of the categories to load to the <tt class="docutils literal"><span class="pre">fetch_20newsgroups</span></tt> function:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="n">cats</span> <span class="o">=</span> <span class="p">[</span><span class="s">'alt.atheism'</span><span class="p">,</span> <span class="s">'sci.space'</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span> <span class="o">=</span> <span class="n">fetch_20newsgroups</span><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s">'train'</span><span class="p">,</span> <span class="n">categories</span><span class="o">=</span><span class="n">cats</span><span class="p">)</span>
<span class="gp">>>> </span><span class="nb">list</span><span class="p">(</span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target_names</span><span class="p">)</span>
<span class="go">['alt.atheism', 'sci.space']</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">filenames</span><span class="o">.</span><span class="n">shape</span>
<span class="go">(1073,)</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target</span><span class="o">.</span><span class="n">shape</span>
<span class="go">(1073,)</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target</span><span class="p">[:</span><span class="mi">10</span><span class="p">]</span>
<span class="go">array([1, 1, 1, 0, 1, 0, 0, 1, 1, 1])</span>
</pre></div>
</div>
</div>
<div class="section" id="converting-text-to-vectors">
<h2>Converting text to vectors<a class="headerlink" href="#converting-text-to-vectors" title="Permalink to this headline">¶</a></h2>
<p>In order to feed predictive or clustering models with the text data,
one first need to turn the text into vectors of numerical values suitable
for statistical analysis. This can be achieved with the utilities of the
<tt class="docutils literal"><span class="pre">sklearn.feature_extraction.text</span></tt> as demonstrated in the following
example that extract <a class="reference external" href="http://en.wikipedia.org/wiki/Tf-idf">TF-IDF</a> vectors of unigram tokens
from a subset of 20news:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.feature_extraction.text</span> <span class="kn">import</span> <span class="n">TfidfVectorizer</span>
<span class="gp">>>> </span><span class="n">categories</span> <span class="o">=</span> <span class="p">[</span><span class="s">'alt.atheism'</span><span class="p">,</span> <span class="s">'talk.religion.misc'</span><span class="p">,</span>
<span class="gp">... </span> <span class="s">'comp.graphics'</span><span class="p">,</span> <span class="s">'sci.space'</span><span class="p">]</span>
<span class="gp">>>> </span><span class="n">newsgroups_train</span> <span class="o">=</span> <span class="n">fetch_20newsgroups</span><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s">'train'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">categories</span><span class="o">=</span><span class="n">categories</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">vectorizer</span> <span class="o">=</span> <span class="n">TfidfVectorizer</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">vectors</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">vectors</span><span class="o">.</span><span class="n">shape</span>
<span class="go">(2034, 34118)</span>
</pre></div>
</div>
<p>The extracted TF-IDF vectors are very sparse, with an average of 159 non-zero
components by sample in a more than 30000-dimensional space
(less than .5% non-zero features):</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="n">vectors</span><span class="o">.</span><span class="n">nnz</span> <span class="o">/</span> <span class="nb">float</span><span class="p">(</span><span class="n">vectors</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
<span class="go">159.01327433628319</span>
</pre></div>
</div>
<p><tt class="docutils literal"><span class="pre">sklearn.datasets.fetch_20newsgroups_vectorized</span></tt> is a function which returns
ready-to-use tfidf features instead of file names.</p>
</div>
<div class="section" id="filtering-text-for-more-realistic-training">
<h2>Filtering text for more realistic training<a class="headerlink" href="#filtering-text-for-more-realistic-training" title="Permalink to this headline">¶</a></h2>
<p>It is easy for a classifier to overfit on particular things that appear in the
20 Newsgroups data, such as newsgroup headers. Many classifiers achieve very
high F-scores, but their results would not generalize to other documents that
aren’t from this window of time.</p>
<p>For example, let’s look at the results of a multinomial Naive Bayes classifier,
which is fast to train and achieves a decent F-score:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.naive_bayes</span> <span class="kn">import</span> <span class="n">MultinomialNB</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn</span> <span class="kn">import</span> <span class="n">metrics</span>
<span class="gp">>>> </span><span class="n">newsgroups_test</span> <span class="o">=</span> <span class="n">fetch_20newsgroups</span><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s">'test'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">categories</span><span class="o">=</span><span class="n">categories</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">vectors_test</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">newsgroups_test</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">MultinomialNB</span><span class="p">(</span><span class="n">alpha</span><span class="o">=.</span><span class="mo">01</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">clf</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">vectors</span><span class="p">,</span> <span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">pred</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">vectors_test</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">(</span><span class="n">newsgroups_test</span><span class="o">.</span><span class="n">target</span><span class="p">,</span> <span class="n">pred</span><span class="p">)</span>
<span class="go">0.88251152461278892</span>
</pre></div>
</div>
<p>(The example <a class="reference internal" href="../auto_examples/document_classification_20newsgroups.html#example-document-classification-20newsgroups-py"><em>Classification of text documents using sparse features</em></a> shuffles
the training and test data, instead of segmenting by time, and in that case
multinomial Naive Bayes gets a much higher F-score of 0.88. Are you suspicious
yet of what’s going on inside this classifier?)</p>
<p>Let’s take a look at what the most informative features are:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">numpy</span> <span class="kn">as</span> <span class="nn">np</span>
<span class="gp">>>> </span><span class="k">def</span> <span class="nf">show_top10</span><span class="p">(</span><span class="n">classifier</span><span class="p">,</span> <span class="n">vectorizer</span><span class="p">,</span> <span class="n">categories</span><span class="p">):</span>
<span class="gp">... </span> <span class="n">feature_names</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">vectorizer</span><span class="o">.</span><span class="n">get_feature_names</span><span class="p">())</span>
<span class="gp">... </span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">category</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">categories</span><span class="p">):</span>
<span class="gp">... </span> <span class="n">top10</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argsort</span><span class="p">(</span><span class="n">classifier</span><span class="o">.</span><span class="n">coef_</span><span class="p">[</span><span class="n">i</span><span class="p">])[</span><span class="o">-</span><span class="mi">10</span><span class="p">:]</span>
<span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="s">"</span><span class="si">%s</span><span class="s">: </span><span class="si">%s</span><span class="s">"</span> <span class="o">%</span> <span class="p">(</span><span class="n">category</span><span class="p">,</span> <span class="s">" "</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">feature_names</span><span class="p">[</span><span class="n">top10</span><span class="p">])))</span>
<span class="gp">...</span>
<span class="gp">>>> </span><span class="n">show_top10</span><span class="p">(</span><span class="n">clf</span><span class="p">,</span> <span class="n">vectorizer</span><span class="p">,</span> <span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target_names</span><span class="p">)</span>
<span class="go">alt.atheism: sgi livesey atheists writes people caltech com god keith edu</span>
<span class="go">comp.graphics: organization thanks files subject com image lines university edu graphics</span>
<span class="go">sci.space: toronto moon gov com alaska access henry nasa edu space</span>
<span class="go">talk.religion.misc: article writes kent people christian jesus sandvik edu com god</span>
</pre></div>
</div>
<p>You can now see many things that these features have overfit to:</p>
<ul class="simple">
<li>Almost every group is distinguished by whether headers such as
<tt class="docutils literal"><span class="pre">NNTP-Posting-Host:</span></tt> and <tt class="docutils literal"><span class="pre">Distribution:</span></tt> appear more or less often.</li>
<li>Another significant feature involves whether the sender is affiliated with
a university, as indicated either by their headers or their signature.</li>
<li>The word “article” is a significant feature, based on how often people quote
previous posts like this: “In article [article ID], [name] <[e-mail address]>
wrote:”</li>
<li>Other features match the names and e-mail addresses of particular people who
were posting at the time.</li>
</ul>
<p>With such an abundance of clues that distinguish newsgroups, the classifiers
barely have to identify topics from text at all, and they all perform at the
same high level.</p>
<p>For this reason, the functions that load 20 Newsgroups data provide a
parameter called <strong>remove</strong>, telling it what kinds of information to strip out
of each file. <strong>remove</strong> should be a tuple containing any subset of
<tt class="docutils literal"><span class="pre">('headers',</span> <span class="pre">'footers',</span> <span class="pre">'quotes')</span></tt>, telling it to remove headers, signature
blocks, and quotation blocks respectively.</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="n">newsgroups_test</span> <span class="o">=</span> <span class="n">fetch_20newsgroups</span><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s">'test'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">remove</span><span class="o">=</span><span class="p">(</span><span class="s">'headers'</span><span class="p">,</span> <span class="s">'footers'</span><span class="p">,</span> <span class="s">'quotes'</span><span class="p">),</span>
<span class="gp">... </span> <span class="n">categories</span><span class="o">=</span><span class="n">categories</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">vectors_test</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">newsgroups_test</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">pred</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">vectors_test</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">(</span><span class="n">pred</span><span class="p">,</span> <span class="n">newsgroups_test</span><span class="o">.</span><span class="n">target</span><span class="p">)</span>
<span class="go">0.78409163025839435</span>
</pre></div>
</div>
<p>This classifier lost over a lot of its F-score, just because we removed
metadata that has little to do with topic classification.
It loses even more if we also strip this metadata from the training data:</p>
<div class="highlight-python"><div class="highlight"><pre><span class="gp">>>> </span><span class="n">newsgroups_train</span> <span class="o">=</span> <span class="n">fetch_20newsgroups</span><span class="p">(</span><span class="n">subset</span><span class="o">=</span><span class="s">'train'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">remove</span><span class="o">=</span><span class="p">(</span><span class="s">'headers'</span><span class="p">,</span> <span class="s">'footers'</span><span class="p">,</span> <span class="s">'quotes'</span><span class="p">),</span>
<span class="gp">... </span> <span class="n">categories</span><span class="o">=</span><span class="n">categories</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">vectors</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">newsgroups_train</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">clf</span> <span class="o">=</span> <span class="n">BernoulliNB</span><span class="p">(</span><span class="n">alpha</span><span class="o">=.</span><span class="mo">01</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">clf</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">vectors</span><span class="p">,</span> <span class="n">newsgroups_train</span><span class="o">.</span><span class="n">target</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">vectors_test</span> <span class="o">=</span> <span class="n">vectorizer</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">newsgroups_test</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">pred</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">vectors_test</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">(</span><span class="n">newsgroups_test</span><span class="o">.</span><span class="n">target</span><span class="p">,</span> <span class="n">pred</span><span class="p">)</span>
<span class="go">0.73160869205141166</span>
</pre></div>
</div>
<p>Some other classifiers cope better with this harder version of the task. Try
running <a class="reference internal" href="../auto_examples/grid_search_text_feature_extraction.html#example-grid-search-text-feature-extraction-py"><em>Sample pipeline for text feature extraction and evaluation</em></a> with and without
the <tt class="docutils literal"><span class="pre">--filter</span></tt> option to compare the results.</p>
<div class="topic">
<p class="topic-title first">Recommendation</p>
<p>When evaluating text classifiers on the 20 Newsgroups data, you
should strip newsgroup-related metadata. In scikit-learn, you can do this by
setting <tt class="docutils literal"><span class="pre">remove=('headers',</span> <span class="pre">'footers',</span> <span class="pre">'quotes')</span></tt>. The F-score will be
lower because it is more realistic.</p>
</div>
<div class="topic">
<p class="topic-title first">Examples</p>
<ul class="simple">
<li><a class="reference internal" href="../auto_examples/grid_search_text_feature_extraction.html#example-grid-search-text-feature-extraction-py"><em>Sample pipeline for text feature extraction and evaluation</em></a></li>
<li><a class="reference internal" href="../auto_examples/document_classification_20newsgroups.html#example-document-classification-20newsgroups-py"><em>Classification of text documents using sparse features</em></a></li>
</ul>
</div>
</div>
</div>
</div>
</div>
</div>
<div class="clearer"></div>
</div>
</div>
<div class="footer">
© 2010 - 2014, scikit-learn developers (BSD License).
<a href="../_sources/datasets/twenty_newsgroups.txt" rel="nofollow">Show this page source</a>
</div>
<div class="rel rellarge">
<div class="buttonPrevious">
<a href="olivetti_faces.html">Previous
</a>
</div>
<div class="buttonNext">
<a href="../modules/classes.html">Next
</a>
</div>
</div>
<script type="text/javascript">
var _gaq = _gaq || [];
_gaq.push(['_setAccount', 'UA-22606712-2']);
_gaq.push(['_trackPageview']);
(function() {
var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
})();
</script>
<script src="http://www.google.com/jsapi" type="text/javascript"></script>
<script type="text/javascript"> google.load('search', '1',
{language : 'en'}); google.setOnLoadCallback(function() {
var customSearchControl = new
google.search.CustomSearchControl('016639176250731907682:tjtqbvtvij0');
customSearchControl.setResultSetSize(google.search.Search.FILTERED_CSE_RESULTSET);
var options = new google.search.DrawOptions();
options.setAutoComplete(true);
customSearchControl.draw('cse', options); }, true);
</script>
<script src="https://scikit-learn.org/versionwarning.js"></script>
</body>
</html>