389:
63:
22:
165:
285:-first principal components in each local neighborhood. It then optimizes to find an embedding that aligns the tangent spaces, but it ignores the label information conveyed by
80:
35:
430:
359:
Ma, L.; Crawford, M. M.; Tian, J. W. (2010). "Generalised supervised local tangent space alignment for hyperspectral image classification".
127:
99:
187:
41:
106:
223:
205:
146:
49:
259:
data, and can also reconstruct high-dimensional coordinates from embedding coordinates. It is based on the intuition that when a
318:
Zhang, Zhenyue; Hongyuan Zha (2004). "Principal
Manifolds and Nonlinear Dimension Reduction via Local Tangent Space Alignment".
113:
84:
423:
271:
95:
449:
183:
73:
459:
416:
120:
327:
175:
388:
332:
286:
454:
244:
400:
368:
337:
256:
248:
443:
278:
62:
396:
341:
267:
252:
372:
260:
264:
298:
270:
to the manifold will become aligned. It begins by computing the
158:
56:
15:
289:, and thus can not be used for classification directly.
404:
87:. Unsourced material may be challenged and removed.
424:
8:
186:. There might be a discussion about this on
50:Learn how and when to remove these messages
431:
417:
331:
224:Learn how and when to remove this message
206:Learn how and when to remove this message
147:Learn how and when to remove this message
310:
7:
385:
383:
320:SIAM Journal on Scientific Computing
85:adding citations to reliable sources
403:. You can help Knowledge (XXG) by
263:is correctly unfolded, all of the
14:
277:of every point. It computes the
31:This article has multiple issues.
387:
281:at every point by computing the
247:, which can efficiently learn a
163:
61:
20:
96:"Local tangent space alignment"
72:needs additional citations for
39:or discuss these issues on the
1:
237:Local tangent space alignment
476:
382:
342:10.1137/s1064827502419154
399:-related article is a
373:10.1049/el.2010.2613
176:confusing or unclear
81:improve this article
450:Dimension reduction
361:Electronics Letters
184:clarify the article
275:-nearest neighbors
243:) is a method for
412:
411:
255:coordinates from
245:manifold learning
234:
233:
226:
216:
215:
208:
157:
156:
149:
131:
54:
467:
460:Statistics stubs
433:
426:
419:
391:
384:
376:
346:
345:
335:
315:
257:high-dimensional
229:
222:
211:
204:
200:
197:
191:
167:
166:
159:
152:
145:
141:
138:
132:
130:
89:
65:
57:
46:
24:
23:
16:
475:
474:
470:
469:
468:
466:
465:
464:
440:
439:
438:
437:
380:
358:
355:
353:Further reading
350:
349:
333:10.1.1.211.9957
317:
316:
312:
307:
295:
253:low-dimensional
251:embedding into
230:
219:
218:
217:
212:
201:
195:
192:
181:
168:
164:
153:
142:
136:
133:
90:
88:
78:
66:
25:
21:
12:
11:
5:
473:
471:
463:
462:
457:
452:
442:
441:
436:
435:
428:
421:
413:
410:
409:
392:
378:
377:
354:
351:
348:
347:
326:(1): 313–338.
309:
308:
306:
303:
302:
301:
294:
291:
232:
231:
214:
213:
196:September 2009
171:
169:
162:
155:
154:
137:September 2009
69:
67:
60:
55:
29:
28:
26:
19:
13:
10:
9:
6:
4:
3:
2:
472:
461:
458:
456:
453:
451:
448:
447:
445:
434:
429:
427:
422:
420:
415:
414:
408:
406:
402:
398:
393:
390:
386:
381:
374:
370:
366:
362:
357:
356:
352:
343:
339:
334:
329:
325:
321:
314:
311:
304:
300:
297:
296:
292:
290:
288:
284:
280:
279:tangent space
276:
274:
269:
266:
262:
258:
254:
250:
246:
242:
238:
228:
225:
210:
207:
199:
189:
188:the talk page
185:
179:
177:
172:This article
170:
161:
160:
151:
148:
140:
129:
126:
122:
119:
115:
112:
108:
105:
101:
98: –
97:
93:
92:Find sources:
86:
82:
76:
75:
70:This article
68:
64:
59:
58:
53:
51:
44:
43:
38:
37:
32:
27:
18:
17:
405:expanding it
394:
379:
364:
360:
323:
319:
313:
287:data samples
282:
272:
240:
236:
235:
220:
202:
193:
182:Please help
173:
143:
134:
124:
117:
110:
103:
91:
79:Please help
74:verification
71:
47:
40:
34:
33:Please help
30:
268:hyperplanes
444:Categories
397:statistics
367:(7): 497.
305:References
178:to readers
107:newspapers
36:improve it
455:Manifolds
328:CiteSeerX
249:nonlinear
42:talk page
293:See also
261:manifold
265:tangent
174:may be
121:scholar
330:
299:Isomap
123:
116:
109:
102:
94:
395:This
128:JSTOR
114:books
401:stub
241:LTSA
100:news
369:doi
338:doi
83:by
446::
365:46
363:.
336:.
324:26
322:.
45:.
432:e
425:t
418:v
407:.
375:.
371::
344:.
340::
283:d
273:k
239:(
227:)
221:(
209:)
203:(
198:)
194:(
190:.
180:.
150:)
144:(
139:)
135:(
125:·
118:·
111:·
104:·
77:.
52:)
48:(
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.