Navegación de contexto

church-corpus-tfidf-lsa.ipynb @ a805aa6

baseconstituyenteestudiantesgeneralplan_patriasala

Last change on this file since a805aa6 was 0ff122b, checked in by rudmanmrrod <rudman22@…>, 7 años ago
Agregado módulo de gestión de perfiles de procesamiento, incorporado el módulo de visualización de modelado de tópicos
Propiedad mode establecida a `100644`
File size: 42.1 KB

Línea
1	{
2	"metadata": {
3	"name": "",
4	"signature": "sha256:1a2c284881d0d090a8144448ddffa2622095c15049758d95804689d0f0432b74"
5	},
6	"nbformat": 3,
7	"nbformat_minor": 0,
8	"worksheets": [
9	{
10	"cells": [
11	{
12	"cell_type": "code",
13	"collapsed": false,
14	"input": [
15	"from vsm import *\n",
16	"from vsm.extensions.corpusbuilders import toy_corpus\n",
17	"\n",
18	"\n",
19	"plain_corpus = \"\"\"\n",
20	"His theology challenged the Pope of the Roman Catholic Church by\n",
21	"teaching that the Bible is the only source of divinely revealed\n",
22	"knowledge.\n",
23	"\n",
24	"Augustine is held in the Catholic Church to be the model teacher.\n",
25	"\n",
26	"Augustine was recognized as a Doctor of the Church by Pope Boniface\n",
27	"VIII.\n",
28	"\n",
29	"Roman Catholic theology stated that faith alone cannot justify man.\n",
30	"\n",
31	"In the Catholic Church the Pope is regarded as the successor of Saint\n",
32	"Peter.\n",
33	"\n",
34	"Alonzo Church was an American mathematician and logician who made\n",
35	"major contributions to mathematical logic and the foundations of\n",
36	"theoretical computer science.\n",
37	"\n",
38	"The lambda calculus was introduced by mathematician Alonzo Church as\n",
39	"an investigation into the foundations of mathematics.\n",
40	"\n",
41	"The Church Turing thesis states that a function is algorithmically\n",
42	"computable if and only if it is computable by a Turing machine.\n",
43	"\n",
44	"Mathematical logic has close connections to the foundations of\n",
45	"mathematics, theoretical computer science.\n",
46	"\n",
47	"A Turing machine can be adapted to simulate the logic of any computer\n",
48	"algorithm.\n",
49	"\"\"\"\n",
50	"\n",
51	"metadata = ['Ecclesiastical ' + str(i) for i in xrange(1, 6)]\n",
52	"metadata += ['Logic ' + str(i) for i in xrange(1, 6)]\n",
53	"\n",
54	"c = toy_corpus(plain_corpus, nltk_stop=True, metadata=metadata)\n",
55	"\n",
56	"tf_m = TfMulti(c, 'document')\n",
57	"tf_m.train(2)\n",
58	"\n",
59	"tfidf_m = TfIdf(tf_m.matrix, 'document')\n",
60	"tfidf_m.train()\n",
61	"\n",
62	"lsa_m = Lsa(tfidf_m.matrix, 'document')\n",
63	"lsa_m.train(k_factors=3)\n",
64	"\n",
65	"tf_v = TfViewer(c, tf_m)\n",
66	"tfidf_v = TfIdfViewer(c, tfidf_m)\n",
67	"lsa_v = LsaViewer(c, lsa_m)"
68	],
69	"language": "python",
70	"metadata": {},
71	"outputs": [
72	{
73	"output_type": "stream",
74	"stream": "stdout",
75	"text": [
76	"Mapping\n"
77	]
78	},
79	{
80	"output_type": "stream",
81	"stream": "stderr",
82	"text": [
83	"/usr/lib/pymodules/python2.7/numpy/ctypeslib.py:411: RuntimeWarning: Item size computed from the PEP 3118 buffer format string does not match the actual item size.\n",
84	" return array(obj, copy=False)\n"
85	]
86	},
87	{
88	"output_type": "stream",
89	"stream": "stdout",
90	"text": [
91	"Reducing\n"
92	]
93	}
94	],
95	"prompt_number": 1
96	},
97	{
98	"cell_type": "code",
99	"collapsed": false,
100	"input": [
101	"tf_v.dist_word_word('logic', print_len=24)"
102	],
103	"language": "python",
104	"metadata": {},
105	"outputs": [
106	{
107	"html": [
108	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"4\">Words: logic</th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>logic </td><td>0.00000 </td><td>close </td><td>0.95532 </td></tr><tr><td>computer </td><td>0.00000 </td><td>adapted </td><td>0.95532 </td></tr><tr><td>theoretical </td><td>0.61548 </td><td>algorithm </td><td>0.95532 </td></tr><tr><td>science </td><td>0.61548 </td><td>american </td><td>0.95532 </td></tr><tr><td>mathematical </td><td>0.61548 </td><td>mathematician </td><td>1.15026 </td></tr><tr><td>foundations </td><td>0.84107 </td><td>alonzo </td><td>1.15026 </td></tr><tr><td>simulate </td><td>0.95532 </td><td>machine </td><td>1.15026 </td></tr><tr><td>major </td><td>0.95532 </td><td>mathematics </td><td>1.15026 </td></tr><tr><td>made </td><td>0.95532 </td><td>turing </td><td>1.30964 </td></tr><tr><td>logician </td><td>0.95532 </td><td>church </td><td>1.35081 </td></tr><tr><td>contributions </td><td>0.95532 </td><td>pope </td><td>1.57080 </td></tr><tr><td>connections </td><td>0.95532 </td><td>recognized </td><td>1.57080 </td></tr></table>"
109	],
110	"metadata": {},
111	"output_type": "pyout",
112	"prompt_number": 2,
113	"text": [
114	"LabeledColumn([('logic', 0.0), ('computer', 0.0),\n",
115	" ('theoretical', 0.6154797086703874),\n",
116	" ('science', 0.6154797086703874),\n",
117	" ('mathematical', 0.6154797086703874),\n",
118	" ('foundations', 0.84106867056793), ('simulate', 0.9553166181245092),\n",
119	" ('major', 0.9553166181245092), ('made', 0.9553166181245092),\n",
120	" ('logician', 0.9553166181245092),\n",
121	" ('contributions', 0.9553166181245092),\n",
122	" ('connections', 0.9553166181245092), ('close', 0.9553166181245092),\n",
123	" ('adapted', 0.9553166181245092), ('algorithm', 0.9553166181245092),\n",
124	" ('american', 0.9553166181245092),\n",
125	" ('mathematician', 1.1502619915109316),\n",
126	" ('alonzo', 1.1502619915109316), ('machine', 1.1502619915109316),\n",
127	" ('mathematics', 1.1502619915109316), ('turing', 1.3096389158918722),\n",
128	" ('church', 1.3508083493994372), ('pope', 1.5707963267948966),\n",
129	" ('recognized', 1.5707963267948966),\n",
130	" ('regarded', 1.5707963267948966), ('revealed', 1.5707963267948966),\n",
131	" ('roman', 1.5707963267948966), ('saint', 1.5707963267948966),\n",
132	" ('algorithmically', 1.5707963267948966),\n",
133	" ('source', 1.5707963267948966), ('stated', 1.5707963267948966),\n",
134	" ('states', 1.5707963267948966), ('successor', 1.5707963267948966),\n",
135	" ('teacher', 1.5707963267948966), ('teaching', 1.5707963267948966),\n",
136	" ('theology', 1.5707963267948966), ('thesis', 1.5707963267948966),\n",
137	" ('alone', 1.5707963267948966), ('peter', 1.5707963267948966),\n",
138	" ('computable', 1.5707963267948966), ('man', 1.5707963267948966),\n",
139	" ('challenged', 1.5707963267948966),\n",
140	" ('catholic', 1.5707963267948966), ('calculus', 1.5707963267948966),\n",
141	" ('divinely', 1.5707963267948966), ('doctor', 1.5707963267948966),\n",
142	" ('faith', 1.5707963267948966), ('boniface', 1.5707963267948966),\n",
143	" ('model', 1.5707963267948966), ('function', 1.5707963267948966),\n",
144	" ('introduced', 1.5707963267948966),\n",
145	" ('investigation', 1.5707963267948966),\n",
146	" ('justify', 1.5707963267948966), ('knowledge', 1.5707963267948966),\n",
147	" ('lambda', 1.5707963267948966), ('bible', 1.5707963267948966),\n",
148	" ('augustine', 1.5707963267948966), ('held', 1.5707963267948966),\n",
149	" ('viii', 1.5707963267948966)], \n",
150	" dtype=[('word', '\|S15'), ('value', '<f8')])"
151	]
152	}
153	],
154	"prompt_number": 2
155	},
156	{
157	"cell_type": "code",
158	"collapsed": false,
159	"input": [
160	"tfidf_v.dist_word_word('logic', print_len=24)"
161	],
162	"language": "python",
163	"metadata": {},
164	"outputs": [
165	{
166	"html": [
167	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"4\">Words: logic</th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>logic </td><td>0.00000 </td><td>close </td><td>0.95532 </td></tr><tr><td>computer </td><td>0.00000 </td><td>adapted </td><td>0.95532 </td></tr><tr><td>theoretical </td><td>0.61548 </td><td>algorithm </td><td>0.95532 </td></tr><tr><td>science </td><td>0.61548 </td><td>american </td><td>0.95532 </td></tr><tr><td>mathematical </td><td>0.61548 </td><td>mathematician </td><td>1.15026 </td></tr><tr><td>foundations </td><td>0.84107 </td><td>alonzo </td><td>1.15026 </td></tr><tr><td>simulate </td><td>0.95532 </td><td>machine </td><td>1.15026 </td></tr><tr><td>major </td><td>0.95532 </td><td>mathematics </td><td>1.15026 </td></tr><tr><td>made </td><td>0.95532 </td><td>turing </td><td>1.30964 </td></tr><tr><td>logician </td><td>0.95532 </td><td>church </td><td>1.35081 </td></tr><tr><td>contributions </td><td>0.95532 </td><td>pope </td><td>1.57080 </td></tr><tr><td>connections </td><td>0.95532 </td><td>recognized </td><td>1.57080 </td></tr></table>"
168	],
169	"metadata": {},
170	"output_type": "pyout",
171	"prompt_number": 3,
172	"text": [
173	"LabeledColumn([('logic', 0.0), ('computer', 0.0),\n",
174	" ('theoretical', 0.6154797086703874),\n",
175	" ('science', 0.6154797086703874),\n",
176	" ('mathematical', 0.6154797086703874),\n",
177	" ('foundations', 0.8410686705679303),\n",
178	" ('simulate', 0.9553166181245094), ('major', 0.9553166181245094),\n",
179	" ('made', 0.9553166181245094), ('logician', 0.9553166181245094),\n",
180	" ('contributions', 0.9553166181245094),\n",
181	" ('connections', 0.9553166181245094), ('close', 0.9553166181245094),\n",
182	" ('adapted', 0.9553166181245094), ('algorithm', 0.9553166181245094),\n",
183	" ('american', 0.9553166181245094),\n",
184	" ('mathematician', 1.1502619915109316),\n",
185	" ('alonzo', 1.1502619915109316), ('machine', 1.1502619915109316),\n",
186	" ('mathematics', 1.1502619915109316), ('turing', 1.3096389158918724),\n",
187	" ('church', 1.3508083493994372), ('pope', 1.5707963267948966),\n",
188	" ('recognized', 1.5707963267948966),\n",
189	" ('regarded', 1.5707963267948966), ('revealed', 1.5707963267948966),\n",
190	" ('roman', 1.5707963267948966), ('saint', 1.5707963267948966),\n",
191	" ('algorithmically', 1.5707963267948966),\n",
192	" ('source', 1.5707963267948966), ('stated', 1.5707963267948966),\n",
193	" ('states', 1.5707963267948966), ('successor', 1.5707963267948966),\n",
194	" ('teacher', 1.5707963267948966), ('teaching', 1.5707963267948966),\n",
195	" ('theology', 1.5707963267948966), ('thesis', 1.5707963267948966),\n",
196	" ('alone', 1.5707963267948966), ('peter', 1.5707963267948966),\n",
197	" ('computable', 1.5707963267948966), ('man', 1.5707963267948966),\n",
198	" ('challenged', 1.5707963267948966),\n",
199	" ('catholic', 1.5707963267948966), ('calculus', 1.5707963267948966),\n",
200	" ('divinely', 1.5707963267948966), ('doctor', 1.5707963267948966),\n",
201	" ('faith', 1.5707963267948966), ('boniface', 1.5707963267948966),\n",
202	" ('model', 1.5707963267948966), ('function', 1.5707963267948966),\n",
203	" ('introduced', 1.5707963267948966),\n",
204	" ('investigation', 1.5707963267948966),\n",
205	" ('justify', 1.5707963267948966), ('knowledge', 1.5707963267948966),\n",
206	" ('lambda', 1.5707963267948966), ('bible', 1.5707963267948966),\n",
207	" ('augustine', 1.5707963267948966), ('held', 1.5707963267948966),\n",
208	" ('viii', 1.5707963267948966)], \n",
209	" dtype=[('word', '\|S15'), ('value', '<f8')])"
210	]
211	}
212	],
213	"prompt_number": 3
214	},
215	{
216	"cell_type": "code",
217	"collapsed": false,
218	"input": [
219	"lsa_v.dist_word_word('logic', print_len=24)"
220	],
221	"language": "python",
222	"metadata": {},
223	"outputs": [
224	{
225	"html": [
226	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"4\">Words: logic</th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>logic </td><td>0.00000 </td><td>contributions </td><td>0.16133 </td></tr><tr><td>computer </td><td>0.00000 </td><td>major </td><td>0.16133 </td></tr><tr><td>connections </td><td>0.14615 </td><td>alonzo </td><td>0.16665 </td></tr><tr><td>close </td><td>0.14615 </td><td>mathematician </td><td>0.16665 </td></tr><tr><td>theoretical </td><td>0.15619 </td><td>calculus </td><td>0.18072 </td></tr><tr><td>science </td><td>0.15619 </td><td>introduced </td><td>0.18072 </td></tr><tr><td>mathematical </td><td>0.15619 </td><td>lambda </td><td>0.18072 </td></tr><tr><td>mathematics </td><td>0.16088 </td><td>investigation </td><td>0.18072 </td></tr><tr><td>foundations </td><td>0.16111 </td><td>church </td><td>0.85232 </td></tr><tr><td>made </td><td>0.16133 </td><td>simulate </td><td>0.94910 </td></tr><tr><td>american </td><td>0.16133 </td><td>adapted </td><td>0.94910 </td></tr><tr><td>logician </td><td>0.16133 </td><td>algorithm </td><td>0.94910 </td></tr></table>"
227	],
228	"metadata": {},
229	"output_type": "pyout",
230	"prompt_number": 4,
231	"text": [
232	"LabeledColumn([('logic', 2.1073424255447017e-08),\n",
233	" ('computer', 2.1073424255447017e-08),\n",
234	" ('connections', 0.1461497327030728), ('close', 0.1461497327030728),\n",
235	" ('theoretical', 0.1561931875925689),\n",
236	" ('science', 0.1561931875925689),\n",
237	" ('mathematical', 0.1561931875925689),\n",
238	" ('mathematics', 0.16088388222038064),\n",
239	" ('foundations', 0.16111028907696195), ('made', 0.16132699126936187),\n",
240	" ('american', 0.16132699126936187),\n",
241	" ('logician', 0.16132699126936187),\n",
242	" ('contributions', 0.16132699126936187),\n",
243	" ('major', 0.16132699126936187), ('alonzo', 0.166652610177689),\n",
244	" ('mathematician', 0.166652610177689),\n",
245	" ('calculus', 0.18071500661099674),\n",
246	" ('introduced', 0.18071500661099674),\n",
247	" ('lambda', 0.18071500661099674),\n",
248	" ('investigation', 0.18071500661099674),\n",
249	" ('church', 0.8523211636647396), ('simulate', 0.9491017050433752),\n",
250	" ('adapted', 0.9491017050433752), ('algorithm', 0.9491017050433752),\n",
251	" ('machine', 1.353933874937494), ('turing', 1.3993762078367922),\n",
252	" ('thesis', 1.455329082734973), ('states', 1.455329082734973),\n",
253	" ('algorithmically', 1.455329082734973),\n",
254	" ('computable', 1.455329082734973), ('function', 1.455329082734973),\n",
255	" ('held', 1.4942581924548715), ('teacher', 1.4942581924548715),\n",
256	" ('model', 1.4942581924548715), ('augustine', 1.5036434979434157),\n",
257	" ('recognized', 1.5100453465502426), ('viii', 1.5100453465502426),\n",
258	" ('boniface', 1.5100453465502426), ('doctor', 1.5100453465502426),\n",
259	" ('peter', 1.537296434294445), ('successor', 1.537296434294445),\n",
260	" ('saint', 1.537296434294445), ('regarded', 1.537296434294445),\n",
261	" ('pope', 1.569538765830574), ('catholic', 1.5741457973418245),\n",
262	" ('challenged', 1.5786643468083752), ('bible', 1.5786643468083752),\n",
263	" ('source', 1.5786643468083752), ('knowledge', 1.5786643468083752),\n",
264	" ('teaching', 1.5786643468083752), ('revealed', 1.5786643468083752),\n",
265	" ('divinely', 1.5786643468083752), ('roman', 1.5807179416878123),\n",
266	" ('theology', 1.5807179416878123), ('alone', 1.5857499354125162),\n",
267	" ('stated', 1.5857499354125162), ('justify', 1.5857499354125162),\n",
268	" ('man', 1.5857499354125162), ('faith', 1.5857499354125162)], \n",
269	" dtype=[('word', '\|S15'), ('value', '<f8')])"
270	]
271	}
272	],
273	"prompt_number": 4
274	},
275	{
276	"cell_type": "code",
277	"collapsed": false,
278	"input": [
279	"tf_v.dist_doc_doc('Ecclesiastical 4')"
280	],
281	"language": "python",
282	"metadata": {},
283	"outputs": [
284	{
285	"html": [
286	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"2\">Documents: </th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Document </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>Ecclesiastical 4 </td><td>0.00000 </td></tr><tr><td>Ecclesiastical 1 </td><td>1.25961 </td></tr><tr><td>Ecclesiastical 2 </td><td>1.42595 </td></tr><tr><td>Ecclesiastical 5 </td><td>1.43676 </td></tr><tr><td>Ecclesiastical 3 </td><td>1.57080 </td></tr><tr><td>Logic 1 </td><td>1.57080 </td></tr><tr><td>Logic 2 </td><td>1.57080 </td></tr><tr><td>Logic 3 </td><td>1.57080 </td></tr><tr><td>Logic 4 </td><td>1.57080 </td></tr><tr><td>Logic 5 </td><td>1.57080 </td></tr></table>"
287	],
288	"metadata": {},
289	"output_type": "pyout",
290	"prompt_number": 5,
291	"text": [
292	"LabeledColumn([('Ecclesiastical 4', 1.4901161193847656e-08),\n",
293	" ('Ecclesiastical 1', 1.2596120825173864),\n",
294	" ('Ecclesiastical 2', 1.4259528297963369),\n",
295	" ('Ecclesiastical 5', 1.4367647653836777),\n",
296	" ('Ecclesiastical 3', 1.5707963267948966),\n",
297	" ('Logic 1', 1.5707963267948966), ('Logic 2', 1.5707963267948966),\n",
298	" ('Logic 3', 1.5707963267948966), ('Logic 4', 1.5707963267948966),\n",
299	" ('Logic 5', 1.5707963267948966)], \n",
300	" dtype=[('doc', '\|S16'), ('value', '<f8')])"
301	]
302	}
303	],
304	"prompt_number": 5
305	},
306	{
307	"cell_type": "code",
308	"collapsed": false,
309	"input": [
310	"tfidf_v.dist_doc_doc('Ecclesiastical 4')"
311	],
312	"language": "python",
313	"metadata": {},
314	"outputs": [
315	{
316	"html": [
317	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"2\">Documents: </th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Document </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>Ecclesiastical 4 </td><td>0.00000 </td></tr><tr><td>Ecclesiastical 1 </td><td>1.41228 </td></tr><tr><td>Ecclesiastical 2 </td><td>1.53742 </td></tr><tr><td>Ecclesiastical 5 </td><td>1.54051 </td></tr><tr><td>Ecclesiastical 3 </td><td>1.57080 </td></tr><tr><td>Logic 1 </td><td>1.57080 </td></tr><tr><td>Logic 2 </td><td>1.57080 </td></tr><tr><td>Logic 3 </td><td>1.57080 </td></tr><tr><td>Logic 4 </td><td>1.57080 </td></tr><tr><td>Logic 5 </td><td>1.57080 </td></tr></table>"
318	],
319	"metadata": {},
320	"output_type": "pyout",
321	"prompt_number": 6,
322	"text": [
323	"LabeledColumn([('Ecclesiastical 4', 0.0), ('Ecclesiastical 1', 1.4122756403782417),\n",
324	" ('Ecclesiastical 2', 1.5374225855153376),\n",
325	" ('Ecclesiastical 5', 1.5405050906164508),\n",
326	" ('Ecclesiastical 3', 1.5707963267948966),\n",
327	" ('Logic 1', 1.5707963267948966), ('Logic 2', 1.5707963267948966),\n",
328	" ('Logic 3', 1.5707963267948966), ('Logic 4', 1.5707963267948966),\n",
329	" ('Logic 5', 1.5707963267948966)], \n",
330	" dtype=[('doc', '\|S16'), ('value', '<f8')])"
331	]
332	}
333	],
334	"prompt_number": 6
335	},
336	{
337	"cell_type": "code",
338	"collapsed": false,
339	"input": [
340	"tfidf_v.dist_word_doc('computer')"
341	],
342	"language": "python",
343	"metadata": {},
344	"outputs": [
345	{
346	"html": [
347	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"2\">Words: </th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Document </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>Logic 5 </td><td>1.32242 </td></tr><tr><td>Logic 4 </td><td>1.32915 </td></tr><tr><td>Logic 1 </td><td>1.38815 </td></tr><tr><td>Ecclesiastical 1 </td><td>1.57080 </td></tr><tr><td>Ecclesiastical 2 </td><td>1.57080 </td></tr><tr><td>Ecclesiastical 3 </td><td>1.57080 </td></tr><tr><td>Ecclesiastical 4 </td><td>1.57080 </td></tr><tr><td>Ecclesiastical 5 </td><td>1.57080 </td></tr><tr><td>Logic 2 </td><td>1.57080 </td></tr><tr><td>Logic 3 </td><td>1.57080 </td></tr></table>"
348	],
349	"metadata": {},
350	"output_type": "pyout",
351	"prompt_number": 7,
352	"text": [
353	"LabeledColumn([('Logic 5', 1.3224154697632784), ('Logic 4', 1.3291532220319409),\n",
354	" ('Logic 1', 1.388146251745384),\n",
355	" ('Ecclesiastical 1', 1.5707963267948966),\n",
356	" ('Ecclesiastical 2', 1.5707963267948966),\n",
357	" ('Ecclesiastical 3', 1.5707963267948966),\n",
358	" ('Ecclesiastical 4', 1.5707963267948966),\n",
359	" ('Ecclesiastical 5', 1.5707963267948966),\n",
360	" ('Logic 2', 1.5707963267948966), ('Logic 3', 1.5707963267948966)], \n",
361	" dtype=[('doc', '\|S16'), ('value', '<f8')])"
362	]
363	}
364	],
365	"prompt_number": 7
366	},
367	{
368	"cell_type": "code",
369	"collapsed": false,
370	"input": [
371	"lsa_v.dist_word_doc('church')"
372	],
373	"language": "python",
374	"metadata": {},
375	"outputs": [
376	{
377	"html": [
378	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"2\">Words: </th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Document </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>Ecclesiastical 2 </td><td>0.73633 </td></tr><tr><td>Ecclesiastical 3 </td><td>0.75591 </td></tr><tr><td>Ecclesiastical 5 </td><td>0.78818 </td></tr><tr><td>Ecclesiastical 1 </td><td>0.83776 </td></tr><tr><td>Logic 5 </td><td>0.84555 </td></tr><tr><td>Ecclesiastical 4 </td><td>0.84617 </td></tr><tr><td>Logic 4 </td><td>0.94199 </td></tr><tr><td>Logic 2 </td><td>0.94433 </td></tr><tr><td>Logic 1 </td><td>0.94556 </td></tr><tr><td>Logic 3 </td><td>1.11314 </td></tr></table>"
379	],
380	"metadata": {},
381	"output_type": "pyout",
382	"prompt_number": 8,
383	"text": [
384	"LabeledColumn([('Ecclesiastical 2', 0.7363264837344443),\n",
385	" ('Ecclesiastical 3', 0.7559069891816871),\n",
386	" ('Ecclesiastical 5', 0.7881792424138612),\n",
387	" ('Ecclesiastical 1', 0.8377568378998803),\n",
388	" ('Logic 5', 0.8455464728606694),\n",
389	" ('Ecclesiastical 4', 0.8461724249009318),\n",
390	" ('Logic 4', 0.9419921003724298), ('Logic 2', 0.9443332777787585),\n",
391	" ('Logic 1', 0.9455622593137681), ('Logic 3', 1.1131440244046271)], \n",
392	" dtype=[('doc', '\|S16'), ('value', '<f8')])"
393	]
394	}
395	],
396	"prompt_number": 8
397	},
398	{
399	"cell_type": "code",
400	"collapsed": false,
401	"input": [
402	"lsa_v.dist_doc_doc('Ecclesiastical 4')"
403	],
404	"language": "python",
405	"metadata": {},
406	"outputs": [
407	{
408	"html": [
409	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"2\">Documents: </th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Document </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>Ecclesiastical 4 </td><td>0.00000 </td></tr><tr><td>Ecclesiastical 1 </td><td>0.00843 </td></tr><tr><td>Ecclesiastical 5 </td><td>0.05804 </td></tr><tr><td>Ecclesiastical 3 </td><td>0.09036 </td></tr><tr><td>Ecclesiastical 2 </td><td>0.10991 </td></tr><tr><td>Logic 2 </td><td>1.56321 </td></tr><tr><td>Logic 3 </td><td>1.57501 </td></tr><tr><td>Logic 1 </td><td>1.57808 </td></tr><tr><td>Logic 4 </td><td>1.58446 </td></tr><tr><td>Logic 5 </td><td>1.58765 </td></tr></table>"
410	],
411	"metadata": {},
412	"output_type": "pyout",
413	"prompt_number": 9,
414	"text": [
415	"LabeledColumn([('Ecclesiastical 4', 0.0),\n",
416	" ('Ecclesiastical 1', 0.008426996338275176),\n",
417	" ('Ecclesiastical 5', 0.05803666882089707),\n",
418	" ('Ecclesiastical 3', 0.09036042875240273),\n",
419	" ('Ecclesiastical 2', 0.10990663361013094),\n",
420	" ('Logic 2', 1.563206166779967), ('Logic 3', 1.5750098468112752),\n",
421	" ('Logic 1', 1.5780801982346484), ('Logic 4', 1.58446244677414),\n",
422	" ('Logic 5', 1.5876493531936278)], \n",
423	" dtype=[('doc', '\|S16'), ('value', '<f8')])"
424	]
425	}
426	],
427	"prompt_number": 9
428	},
429	{
430	"cell_type": "code",
431	"collapsed": false,
432	"input": [
433	"tf_v.coll_freq('church')"
434	],
435	"language": "python",
436	"metadata": {},
437	"outputs": [
438	{
439	"metadata": {},
440	"output_type": "pyout",
441	"prompt_number": 10,
442	"text": [
443	"7"
444	]
445	}
446	],
447	"prompt_number": 10
448	},
449	{
450	"cell_type": "code",
451	"collapsed": false,
452	"input": [
453	"tf_v.coll_freqs(print_len=54)"
454	],
455	"language": "python",
456	"metadata": {},
457	"outputs": [
458	{
459	"html": [
460	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"8\">Collection Frequencies</th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Counts </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Counts </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Counts </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Counts </th></tr><tr><td>church </td><td>7 </td><td>computable </td><td>2 </td><td>calculus </td><td>1 </td><td>states </td><td>1 </td></tr><tr><td>catholic </td><td>4 </td><td>augustine </td><td>2 </td><td>american </td><td>1 </td><td>stated </td><td>1 </td></tr><tr><td>foundations </td><td>3 </td><td>roman </td><td>2 </td><td>close </td><td>1 </td><td>source </td><td>1 </td></tr><tr><td>turing </td><td>3 </td><td>science </td><td>2 </td><td>algorithmically </td><td>1 </td><td>simulate </td><td>1 </td></tr><tr><td>computer </td><td>3 </td><td>alone </td><td>1 </td><td>held </td><td>1 </td><td>saint </td><td>1 </td></tr><tr><td>pope </td><td>3 </td><td>contributions </td><td>1 </td><td>function </td><td>1 </td><td>revealed </td><td>1 </td></tr><tr><td>logic </td><td>3 </td><td>algorithm </td><td>1 </td><td>viii </td><td>1 </td><td>regarded </td><td>1 </td></tr><tr><td>alonzo </td><td>2 </td><td>faith </td><td>1 </td><td>introduced </td><td>1 </td><td>recognized </td><td>1 </td></tr><tr><td>mathematics </td><td>2 </td><td>doctor </td><td>1 </td><td>investigation </td><td>1 </td><td>peter </td><td>1 </td></tr><tr><td>theoretical </td><td>2 </td><td>divinely </td><td>1 </td><td>thesis </td><td>1 </td><td>model </td><td>1 </td></tr><tr><td>machine </td><td>2 </td><td>bible </td><td>1 </td><td>teaching </td><td>1 </td><td>man </td><td>1 </td></tr><tr><td>theology </td><td>2 </td><td>boniface </td><td>1 </td><td>teacher </td><td>1 </td><td>major </td><td>1 </td></tr><tr><td>mathematical </td><td>2 </td><td>connections </td><td>1 </td><td>successor </td><td>1 </td><td>made </td><td>1 </td></tr><tr><td>mathematician </td><td>2 </td><td>challenged </td><td>1 </td><td style=\"border-color: #EFF2FB; background: #EEF2FB;\"> </td><td style=\"border-color: #EFF2FB; background: #EEF2FB;\"> </td><td style=\"border-color: #EFF2FB; background: #EEF2FB;\"> </td><td style=\"border-color: #EFF2FB; background: #EEF2FB;\"> </td></tr></table>"
461	],
462	"metadata": {},
463	"output_type": "pyout",
464	"prompt_number": 11,
465	"text": [
466	"LabeledColumn([('church', 7), ('catholic', 4), ('foundations', 3), ('turing', 3),\n",
467	" ('computer', 3), ('pope', 3), ('logic', 3), ('alonzo', 2),\n",
468	" ('mathematics', 2), ('theoretical', 2), ('machine', 2),\n",
469	" ('theology', 2), ('mathematical', 2), ('mathematician', 2),\n",
470	" ('computable', 2), ('augustine', 2), ('roman', 2), ('science', 2),\n",
471	" ('alone', 1), ('contributions', 1), ('algorithm', 1), ('faith', 1),\n",
472	" ('doctor', 1), ('divinely', 1), ('bible', 1), ('boniface', 1),\n",
473	" ('connections', 1), ('challenged', 1), ('calculus', 1),\n",
474	" ('american', 1), ('close', 1), ('algorithmically', 1), ('held', 1),\n",
475	" ('function', 1), ('viii', 1), ('introduced', 1),\n",
476	" ('investigation', 1), ('thesis', 1), ('teaching', 1),\n",
477	" ('teacher', 1), ('successor', 1), ('states', 1), ('stated', 1),\n",
478	" ('source', 1), ('simulate', 1), ('saint', 1), ('revealed', 1),\n",
479	" ('regarded', 1), ('recognized', 1), ('peter', 1), ('model', 1),\n",
480	" ('man', 1), ('major', 1), ('made', 1), ('logician', 1),\n",
481	" ('lambda', 1), ('knowledge', 1), ('justify', 1), ('adapted', 1)], \n",
482	" dtype=[('word', '\|S15'), ('value', '<i4')])"
483	]
484	}
485	],
486	"prompt_number": 11
487	},
488	{
489	"cell_type": "code",
490	"collapsed": false,
491	"input": [
492	"dm = tf_v.dismat_word(['logic','church','catholic','pope'])\n",
493	"np.around(dm, decimals=2)"
494	],
495	"language": "python",
496	"metadata": {},
497	"outputs": [
498	{
499	"metadata": {},
500	"output_type": "pyout",
501	"prompt_number": 12,
502	"text": [
503	"array([[ 0. , 1.35, 1.57, 1.57],\n",
504	" [ 1.35, 0. , 0.97, 0.86],\n",
505	" [ 1.57, 0.97, 0. , 0.96],\n",
506	" [ 1.57, 0.86, 0.96, 0. ]])"
507	]
508	}
509	],
510	"prompt_number": 12
511	},
512	{
513	"cell_type": "code",
514	"collapsed": false,
515	"input": [
516	"dm = tfidf_v.dismat_word(['logic','church','catholic','pope'])\n",
517	"np.around(dm, decimals=2)"
518	],
519	"language": "python",
520	"metadata": {},
521	"outputs": [
522	{
523	"metadata": {},
524	"output_type": "pyout",
525	"prompt_number": 13,
526	"text": [
527	"array([[ 0. , 1.35, 1.57, 1.57],\n",
528	" [ 1.35, 0. , 0.97, 0.86],\n",
529	" [ 1.57, 0.97, 0. , 0.96],\n",
530	" [ 1.57, 0.86, 0.96, 0. ]])"
531	]
532	}
533	],
534	"prompt_number": 13
535	},
536	{
537	"cell_type": "code",
538	"collapsed": false,
539	"input": [
540	"dm = lsa_v.dismat_word(['logic','church','catholic','pope'])\n",
541	"np.around(dm, decimals=2)"
542	],
543	"language": "python",
544	"metadata": {},
545	"outputs": [
546	{
547	"metadata": {},
548	"output_type": "pyout",
549	"prompt_number": 14,
550	"text": [
551	"array([[ 0. , 0.85, 1.57, 1.57],\n",
552	" [ 0.85, 0. , 0.87, 0.87],\n",
553	" [ 1.57, 0.87, 0. , 0.01],\n",
554	" [ 1.57, 0.87, 0.01, 0. ]])"
555	]
556	}
557	],
558	"prompt_number": 14
559	},
560	{
561	"cell_type": "code",
562	"collapsed": false,
563	"input": [
564	"dm = tf_v.dismat_doc(['Logic 1','Logic 2','Ecclesiastical 1','Ecclesiastical 2'])\n",
565	"np.around(dm, decimals=2)"
566	],
567	"language": "python",
568	"metadata": {},
569	"outputs": [
570	{
571	"metadata": {},
572	"output_type": "pyout",
573	"prompt_number": 15,
574	"text": [
575	"array([[ 0. , 1.21, 1.49, 1.46],\n",
576	" [ 1.21, 0. , 1.47, 1.43],\n",
577	" [ 1.49, 1.47, 0. , 1.33],\n",
578	" [ 1.46, 1.43, 1.33, 0. ]])"
579	]
580	}
581	],
582	"prompt_number": 15
583	},
584	{
585	"cell_type": "code",
586	"collapsed": false,
587	"input": [
588	"dm = tfidf_v.dismat_doc(['Logic 1','Logic 2','Ecclesiastical 1','Ecclesiastical 2'])\n",
589	"np.around(dm, decimals=2)"
590	],
591	"language": "python",
592	"metadata": {},
593	"outputs": [
594	{
595	"metadata": {},
596	"output_type": "pyout",
597	"prompt_number": 16,
598	"text": [
599	"array([[ 0. , 1.39, 1.57, 1.57],\n",
600	" [ 1.39, 0. , 1.57, 1.57],\n",
601	" [ 1.57, 1.57, 0. , 1.54],\n",
602	" [ 1.57, 1.57, 1.54, 0. ]])"
603	]
604	}
605	],
606	"prompt_number": 16
607	},
608	{
609	"cell_type": "code",
610	"collapsed": false,
611	"input": [
612	"dm = lsa_v.dismat_doc(['Logic 1','Logic 2','Ecclesiastical 1','Ecclesiastical 2'])\n",
613	"np.around(dm, decimals=2)"
614	],
615	"language": "python",
616	"metadata": {},
617	"outputs": [
618	{
619	"metadata": {},
620	"output_type": "pyout",
621	"prompt_number": 17,
622	"text": [
623	"array([[ 0. , 0.02, 1.57, 1.49],\n",
624	" [ 0.02, 0. , 1.56, 1.48],\n",
625	" [ 1.57, 1.56, 0. , 0.1 ],\n",
626	" [ 1.49, 1.48, 0.1 , 0. ]])"
627	]
628	}
629	],
630	"prompt_number": 17
631	},
632	{
633	"cell_type": "code",
634	"collapsed": false,
635	"input": [
636	"from vsm.spatial import scipy_cdist\n",
637	"lsa_v.dist_word_word('logic', print_len=24, dist_fn=scipy_cdist(metric='minkowski'))"
638	],
639	"language": "python",
640	"metadata": {},
641	"outputs": [
642	{
643	"html": [
644	"<table style=\"margin: 0\"><tr><th style=\"text-align: center; background: #CEE3F6\" colspan =\"4\">Words: logic</th></tr><tr><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th><th style=\"text-align: center; background: #EFF2FB; \">Word </th><th style=\"text-align: center; background: #EFF2FB; \">Distance </th></tr><tr><td>logic </td><td>0.00000 </td><td>theoretical </td><td>0.06325 </td></tr><tr><td>computer </td><td>0.00000 </td><td>mathematics </td><td>0.07048 </td></tr><tr><td>mathematician </td><td>0.04921 </td><td>connections </td><td>0.09843 </td></tr><tr><td>alonzo </td><td>0.04921 </td><td>close </td><td>0.09843 </td></tr><tr><td>foundations </td><td>0.05166 </td><td>introduced </td><td>0.12865 </td></tr><tr><td>american </td><td>0.05320 </td><td>lambda </td><td>0.12865 </td></tr><tr><td>contributions </td><td>0.05320 </td><td>investigation </td><td>0.12865 </td></tr><tr><td>logician </td><td>0.05320 </td><td>calculus </td><td>0.12865 </td></tr><tr><td>made </td><td>0.05320 </td><td>church </td><td>0.18155 </td></tr><tr><td>major </td><td>0.05320 </td><td>simulate </td><td>0.19473 </td></tr><tr><td>mathematical </td><td>0.06325 </td><td>adapted </td><td>0.19473 </td></tr><tr><td>science </td><td>0.06325 </td><td>algorithm </td><td>0.19473 </td></tr></table>"
645	],
646	"metadata": {},
647	"output_type": "pyout",
648	"prompt_number": 18,
649	"text": [
650	"LabeledColumn([('logic', 0.0), ('computer', 0.0),\n",
651	" ('mathematician', 0.0492069763182823),\n",
652	" ('alonzo', 0.0492069763182823),\n",
653	" ('foundations', 0.05166368875941544),\n",
654	" ('american', 0.05320430492811313),\n",
655	" ('contributions', 0.05320430492811313),\n",
656	" ('logician', 0.05320430492811313), ('made', 0.05320430492811313),\n",
657	" ('major', 0.05320430492811313),\n",
658	" ('mathematical', 0.0632503959697633),\n",
659	" ('science', 0.0632503959697633),\n",
660	" ('theoretical', 0.0632503959697633),\n",
661	" ('mathematics', 0.07048230197449873),\n",
662	" ('connections', 0.09842980401516312),\n",
663	" ('close', 0.09842980401516312), ('introduced', 0.12864998450849807),\n",
664	" ('lambda', 0.12864998450849807),\n",
665	" ('investigation', 0.12864998450849807),\n",
666	" ('calculus', 0.12864998450849807), ('church', 0.18154738827057404),\n",
667	" ('simulate', 0.19473020810363867), ('adapted', 0.19473020810363867),\n",
668	" ('algorithm', 0.19473020810363867), ('held', 0.22833958546575822),\n",
669	" ('teacher', 0.22833958546575822), ('model', 0.22833958546575822),\n",
670	" ('recognized', 0.22893465574112978),\n",
671	" ('boniface', 0.22893465574112978), ('viii', 0.22893465574112978),\n",
672	" ('doctor', 0.22893465574112978), ('augustine', 0.2290393895483237),\n",
673	" ('peter', 0.23082222876942637), ('saint', 0.23082222876942637),\n",
674	" ('successor', 0.23082222876942637),\n",
675	" ('regarded', 0.23082222876942637), ('alone', 0.26236426820101866),\n",
676	" ('justify', 0.26236426820101866), ('stated', 0.26236426820101866),\n",
677	" ('faith', 0.26236426820101866), ('man', 0.26236426820101866),\n",
678	" ('pope', 0.2993972085117987), ('catholic', 0.30013287605358824),\n",
679	" ('machine', 0.3040698811395842),\n",
680	" ('algorithmically', 0.35328304712159975),\n",
681	" ('function', 0.35328304712159975), ('states', 0.35328304712159975),\n",
682	" ('thesis', 0.35328304712159975), ('theology', 0.3796994235501325),\n",
683	" ('roman', 0.3796994235501325), ('revealed', 0.38318073578287704),\n",
684	" ('knowledge', 0.38318073578287704),\n",
685	" ('teaching', 0.38318073578287704), ('bible', 0.38318073578287704),\n",
686	" ('divinely', 0.38318073578287704),\n",
687	" ('challenged', 0.38318073578287704),\n",
688	" ('source', 0.38318073578287704), ('turing', 0.47942958346148457),\n",
689	" ('computable', 0.6109097308593834)], \n",
690	" dtype=[('word', '\|S15'), ('value', '<f8')])"
691	]
692	}
693	],
694	"prompt_number": 18
695	},
696	{
697	"cell_type": "code",
698	"collapsed": false,
699	"input": [
700	"dm = lsa_v.dismat_doc(['Logic 1','Logic 2','Ecclesiastical 1','Ecclesiastical 2'], dist_fn=scipy_cdist(metric='cosine'))\n",
701	"np.around(dm, decimals=2)"
702	],
703	"language": "python",
704	"metadata": {},
705	"outputs": [
706	{
707	"metadata": {},
708	"output_type": "pyout",
709	"prompt_number": 19,
710	"text": [
711	"array([[ 0. , 0. , 1. , 0.92],\n",
712	" [ 0. , 0. , 0.99, 0.91],\n",
713	" [ 1. , 0.99, 0. , 0.01],\n",
714	" [ 0.92, 0.91, 0.01, 0. ]])"
715	]
716	}
717	],
718	"prompt_number": 19
719	}
720	],
721	"metadata": {}
722	}
723	]
724	}

Nota: Vea TracBrowser para ayuda de uso del navegador del repositorio.

Descargar en otros formatos:

Formato original