1022:
1587:. Batch learning algorithms require all the data samples to be available beforehand. It trains the model using the entire training data and then predicts the test sample using the found relationship. The online learning algorithms, on the other hand, incrementally build their models in sequential iterations. In iteration t, an online algorithm receives a sample, x
1560:(MEP) is an evolutionary algorithm for generating computer programs (that can be used for classification tasks too). MEP has a unique feature: it encodes multiple programs into a single chromosome. Each of these programs can be used to generate the output for a class, thus making MEP naturally suitable for solving multi-class classification problems.
990:). For example, deciding on whether an image is showing a banana, an orange, or an apple is a multiclass classification problem, with three possible classes (banana, orange, apple), while deciding on whether an image contains an apple or not is a binary classification problem (with the two possible classes being: apple, no apple).
1548:
are based upon the idea of maximizing the margin i.e. maximizing the minimum distance from the separating hyperplane to the nearest example. The basic SVM supports only binary classification, but extensions have been proposed to handle the multiclass classification case as well. In these extensions,
1508:
kNN is considered among the oldest non-parametric classification algorithms. To classify an unknown example, the distance from that example to every other training example is measured. The k smallest distances are identified, and the most represented class by these k nearest neighbours is considered
1398:
that suffers from several problems. Firstly, the scale of the confidence values may differ between the binary classifiers. Second, even if the class distribution is balanced in the training set, the binary classification learners see unbalanced distributions because typically the set of negatives
1536:
is a powerful classification technique. The tree tries to infer a split of the training data based on the values of the available features to produce a good generalization. The algorithm can naturally handle binary or multiclass classification problems. The leaf nodes can refer to any of the K
1480:
Multiclass perceptrons provide a natural extension to the multi-class problem. Instead of just having one neuron in the output layer, with binary output, one could have N binary neurons leading to multi-class classification. In practice, the last layer of a neural network is usually a
1611:). Recently, a new learning paradigm called progressive learning technique has been developed. The progressive learning technique is capable of not only learning from new samples but also capable of learning new classes of data and yet retain the knowledge learnt thus far.
1485:
layer, which is the algebraic simplification of N logistic classifiers, normalized per class by the sum of the N-1 other logistic classifiers. Neural
Network-based classification has brought significant improvements and scopes for thinking from different perspectives.
1117:, OAA) strategy involves training a single classifier per class, with the samples of that class as positive samples and all other samples as negatives. This strategy requires the base classifiers to produce a real-valued score for its decision (see also
1520:
is a successful classifier based upon the principle of maximum a posteriori (MAP). This approach is naturally extensible to the case of having more than two classes, and was shown to perform well in spite of the underlying simplifying assumption of
1497:(ELM) is a special case of single hidden layer feed-forward neural networks (SLFNs) wherein the input weights and the hidden node biases can be chosen at random. Many variants and developments are made to the ELM for multiclass classification.
1389:
1575:. Each parent node is divided into multiple child nodes and the process is continued until each child node represents only one class. Several methods have been proposed based on hierarchical classification.
1426:-way multiclass problem; each receives the samples of a pair of classes from the original training set, and must learn to distinguish these two classes. At prediction time, a voting scheme is applied: all
1619:
The performance of a multi-class classification system is often assessed by comparing the predictions of the system against reference labels with an evaluation metric. Common evaluation metrics are
884:
922:
879:
869:
1448:
This section discusses strategies of extending the existing binary classifiers to solve multi-class classification problems. Several algorithms have been developed based on
710:
917:
1093:
This section discusses strategies for reducing the problem of multiclass classification to multiple binary classification problems. It can be categorized into
1008:, where multiple labels are to be predicted for each instance (e.g., predicting that an image contains both an apple and an orange, in the previous example).
874:
725:
456:
1306:
1101:. The techniques developed based on reducing the multi-class problem into multiple binary problems can also be called problem transformation techniques.
1039:
957:
760:
1437:
classifiers are applied to an unseen sample and the class that got the highest number of "+1" predictions gets predicted by the combined classifier.
1705:
1121:), rather than just a class label; discrete class labels alone can lead to ambiguities, where multiple classes are predicted for a single sample.
836:
385:
894:
657:
192:
1895:
912:
1890:
745:
720:
669:
1061:
793:
788:
441:
986:
is the problem of classifying instances into one of three or more classes (classifying instances into one of two classes is called
451:
89:
1549:
additional parameters and constraints are added to the optimization problem to handle the separation of the different classes.
994:
1472:
to address multi-class classification problems. These types of techniques can also be called algorithm adaptation techniques.
846:
1043:
950:
610:
431:
1583:
Based on learning paradigms, the existing multi-class classification techniques can be classified into batch learning and
821:
523:
299:
1557:
778:
715:
625:
603:
446:
436:
1829:
1568:
929:
841:
826:
287:
109:
1783:
Venkatesan, Rajasekar; Meng Joo, Er (2016). "A novel progressive learning technique for multi-class classification".
1440:
Like OvR, OvO suffers from ambiguities in that some regions of its input space may receive the same number of votes.
816:
1676:
1646:
1005:
975:
889:
566:
461:
249:
182:
142:
31:
1032:
943:
549:
317:
187:
1641:
1522:
1494:
1469:
571:
491:
414:
332:
162:
124:
119:
79:
74:
1845:"A Closer Look at Classification Evaluation Metrics and a Critical Reflection of Common Evaluation Practice"
1584:
1545:
1465:
518:
367:
267:
94:
1620:
1533:
1124:
In pseudocode, the training algorithm for an OvR learner constructed from a binary classification learner
698:
674:
576:
337:
312:
272:
84:
1636:
1572:
998:
987:
652:
474:
426:
282:
197:
69:
581:
531:
1656:
1505:
1457:
1001:
algorithms; these can, however, be turned into multinomial classifiers by a variety of strategies.
684:
620:
591:
496:
322:
255:
241:
227:
202:
152:
104:
64:
1856:
1810:
1792:
1763:
1762:
Kabir, H M Dipu (2023). "Reduction of class activation uncertainty with background information".
662:
586:
372:
167:
755:
598:
511:
307:
277:
222:
217:
172:
114:
1866:
1802:
1482:
971:
783:
536:
486:
396:
380:
350:
212:
207:
157:
147:
45:
1449:
811:
615:
481:
421:
1571:
tackles the multi-class classification problem by dividing the output space i.e. into a
1453:
831:
362:
99:
1884:
750:
679:
561:
292:
177:
1814:
1384:{\displaystyle {\hat {y}}={\underset {k\in \{1\ldots K\}}{\arg \!\max }}\;f_{k}(x)}
1118:
17:
1806:
1749:
Proceedings of the IEEE/CVF Conference on
Computer Vision and Pattern Recognition
1517:
1461:
1021:
556:
50:
1747:
Ekin, Cubuk (2019). "Autoaugment: Learning augmentation strategies from data".
1683:
and the prediction of multiple classes is considered a feature, not a problem.
1651:
705:
401:
327:
1300:
for which the corresponding classifier reports the highest confidence score:
1395:
864:
645:
1074:
The existing multi-class classification techniques can be categorised into
1870:
997:) naturally permit the use of more than two classes, some are by nature
1844:
1624:
1046: in this section. Unsourced material may be challenged and removed.
640:
391:
1292:
Making decisions means applying all classifiers to an unseen sample
1861:
1797:
1768:
635:
630:
357:
1015:
1849:
Transactions of the
Association for Computational Linguistics
27:
Problem in machine learning and statistical classification
1603:
and updates its model based on the sample-label pair: (x
923:
List of datasets in computer vision and image processing
1138:, a learner (training algorithm for binary classifiers)
1595:
using the current model; the algorithm then receives y
1004:
Multiclass classification should not be confused with
1309:
1399:they see is much larger than the set of positives.
1383:
1331:
1332:
993:While many classification algorithms (notably
918:List of datasets for machine-learning research
1706:"Survey on multiclass classification methods"
951:
8:
1355:
1343:
1394:Although this strategy is popular, it is a
1361:
958:
944:
36:
1860:
1796:
1767:
1366:
1325:
1311:
1310:
1308:
1062:Learn how and when to remove this message
1734:Pattern Recognition and Machine Learning
1727:
1725:
1723:
1721:
1719:
1696:
1668:
44:
7:
1044:adding citations to reliable sources
913:Glossary of artificial intelligence
25:
1830:"Progressive Learning Technique"
1020:
1732:Bishop, Christopher M. (2006).
1031:needs additional citations for
995:multinomial logistic regression
1378:
1372:
1316:
1162:} is the label for the sample
333:Relevance vector machine (RVM)
1:
1210:Construct a new label vector
822:Computational learning theory
386:Expectationâmaximization (EM)
1807:10.1016/j.neucom.2016.05.006
1558:Multi expression programming
1553:Multi expression programming
1411:(OvO) reduction, one trains
1084:hierarchical classification.
779:Coefficient of determination
626:Convolutional neural network
338:Support vector machine (SVM)
1569:Hierarchical classification
1564:Hierarchical classification
930:Outline of machine learning
827:Empirical risk minimization
1912:
1896:Statistical classification
1677:multi-label classification
1647:Multi-label classification
1006:multi-label classification
984:multinomial classification
976:statistical classification
567:Feedforward neural network
318:Artificial neural networks
32:multi-label classification
29:
1891:Classification algorithms
1710:Technical Report, Caltech
1495:Extreme learning machines
1490:Extreme learning machines
1470:extreme learning machines
1422:binary classifiers for a
1296:and predicting the label
980:multiclass classification
550:Artificial neural network
1642:One-class classification
1591:and predicts its label š
1523:conditional independence
1509:the output class label.
1089:Transformation to binary
1078:transformation to binary
859:Journals and conferences
806:Mathematical foundations
716:Temporal difference (TD)
572:Recurrent neural network
492:Conditional random field
415:Dimensionality reduction
163:Dimensionality reduction
125:Quantum machine learning
120:Neuromorphic engineering
80:Self-supervised learning
75:Semi-supervised learning
30:Not to be confused with
1828:Venkatesan, Rajasekar.
1546:Support vector machines
1541:Support vector machines
1466:support vector machines
268:Apprenticeship learning
1534:Decision tree learning
1385:
1176:a list of classifiers
817:Biasâvariance tradeoff
699:Reinforcement learning
675:Spiking neural network
85:Reinforcement learning
1704:Mohamed, Aly (2005).
1652:Multiclass perceptron
1637:Binary classification
1599:, the true label of x
1444:Extension from binary
1386:
1109:One-vs.-rest (OvR or
1081:extension from binary
988:binary classification
653:Neural radiance field
475:Structured prediction
198:Structured prediction
70:Unsupervised learning
1871:10.1162/tacl_a_00675
1843:Opitz, Juri (2024).
1501:k-nearest neighbours
1307:
1040:improve this article
842:Statistical learning
740:Learning with humans
532:Local outlier factor
1657:Multi-task learning
1537:classes concerned.
1506:k-nearest neighbors
1458:k-nearest neighbors
685:Electrochemical RAM
592:reservoir computing
323:Logistic regression
242:Supervised learning
228:Multimodal learning
203:Feature engineering
148:Generative modeling
110:Rule-based learning
105:Curriculum learning
65:Supervised learning
40:Part of a series on
1679:, OvR is known as
1579:Learning paradigms
1381:
1359:
1012:General strategies
253: •
168:Density estimation
18:Multiclass problem
1326:
1319:
1072:
1071:
1064:
968:
967:
773:Model diagnostics
756:Human-in-the-loop
599:Boltzmann machine
512:Anomaly detection
308:Linear regression
223:Ontology learning
218:Grammar induction
193:Semantic analysis
188:Association rules
173:Anomaly detection
115:Neuro-symbolic AI
16:(Redirected from
1903:
1875:
1874:
1864:
1840:
1834:
1833:
1825:
1819:
1818:
1800:
1780:
1774:
1773:
1771:
1759:
1753:
1752:
1744:
1738:
1737:
1729:
1714:
1713:
1701:
1684:
1681:binary relevance
1673:
1483:softmax function
1436:
1425:
1421:
1390:
1388:
1387:
1382:
1371:
1370:
1360:
1358:
1335:
1321:
1320:
1312:
1299:
1295:
1284:
1277:
1273:
1269:
1262:
1250:
1235:
1224:
1213:
1206:
1202:
1190:
1186:
1182:
1168:
1161:
1157:
1150:
1144:
1137:
1127:
1067:
1060:
1056:
1053:
1047:
1024:
1016:
972:machine learning
960:
953:
946:
907:Related articles
784:Confusion matrix
537:Isolation forest
482:Graphical models
261:
260:
213:Learning to rank
208:Feature learning
46:Machine learning
37:
21:
1911:
1910:
1906:
1905:
1904:
1902:
1901:
1900:
1881:
1880:
1879:
1878:
1842:
1841:
1837:
1827:
1826:
1822:
1782:
1781:
1777:
1761:
1760:
1756:
1746:
1745:
1741:
1731:
1730:
1717:
1703:
1702:
1698:
1693:
1688:
1687:
1674:
1670:
1665:
1633:
1617:
1610:
1606:
1602:
1598:
1594:
1590:
1585:online learning
1581:
1566:
1555:
1543:
1531:
1515:
1503:
1492:
1478:
1476:Neural networks
1450:neural networks
1446:
1427:
1423:
1412:
1405:
1362:
1336:
1327:
1305:
1304:
1297:
1293:
1283:
1279:
1275:
1271:
1267:
1260:
1252:
1245:
1237:
1234:
1226:
1223:
1215:
1211:
1204:
1200:
1188:
1184:
1181:
1177:
1167:
1163:
1159:
1156:
1152:
1148:
1142:
1135:
1128:is as follows:
1125:
1115:one-against-all
1107:
1091:
1068:
1057:
1051:
1048:
1037:
1025:
1014:
964:
935:
934:
908:
900:
899:
860:
852:
851:
812:Kernel machines
807:
799:
798:
774:
766:
765:
746:Active learning
741:
733:
732:
701:
691:
690:
616:Diffusion model
552:
542:
541:
514:
504:
503:
477:
467:
466:
422:Factor analysis
417:
407:
406:
390:
353:
343:
342:
263:
262:
246:
245:
244:
233:
232:
138:
130:
129:
95:Online learning
60:
48:
35:
28:
23:
22:
15:
12:
11:
5:
1909:
1907:
1899:
1898:
1893:
1883:
1882:
1877:
1876:
1835:
1820:
1785:Neurocomputing
1775:
1754:
1739:
1715:
1695:
1694:
1692:
1689:
1686:
1685:
1667:
1666:
1664:
1661:
1660:
1659:
1654:
1649:
1644:
1639:
1632:
1629:
1616:
1613:
1608:
1604:
1600:
1596:
1592:
1588:
1580:
1577:
1565:
1562:
1554:
1551:
1542:
1539:
1530:
1529:Decision trees
1527:
1514:
1511:
1502:
1499:
1491:
1488:
1477:
1474:
1454:decision trees
1445:
1442:
1404:
1401:
1392:
1391:
1380:
1377:
1374:
1369:
1365:
1357:
1354:
1351:
1348:
1345:
1342:
1339:
1334:
1330:
1324:
1318:
1315:
1290:
1289:
1288:
1287:
1286:
1285:
1281:
1264:
1256:
1241:
1230:
1219:
1194:
1193:
1192:
1179:
1171:
1170:
1169:
1165:
1154:
1145:
1139:
1106:
1103:
1090:
1087:
1086:
1085:
1082:
1079:
1070:
1069:
1028:
1026:
1019:
1013:
1010:
966:
965:
963:
962:
955:
948:
940:
937:
936:
933:
932:
927:
926:
925:
915:
909:
906:
905:
902:
901:
898:
897:
892:
887:
882:
877:
872:
867:
861:
858:
857:
854:
853:
850:
849:
844:
839:
834:
832:Occam learning
829:
824:
819:
814:
808:
805:
804:
801:
800:
797:
796:
791:
789:Learning curve
786:
781:
775:
772:
771:
768:
767:
764:
763:
758:
753:
748:
742:
739:
738:
735:
734:
731:
730:
729:
728:
718:
713:
708:
702:
697:
696:
693:
692:
689:
688:
682:
677:
672:
667:
666:
665:
655:
650:
649:
648:
643:
638:
633:
623:
618:
613:
608:
607:
606:
596:
595:
594:
589:
584:
579:
569:
564:
559:
553:
548:
547:
544:
543:
540:
539:
534:
529:
521:
515:
510:
509:
506:
505:
502:
501:
500:
499:
494:
489:
478:
473:
472:
469:
468:
465:
464:
459:
454:
449:
444:
439:
434:
429:
424:
418:
413:
412:
409:
408:
405:
404:
399:
394:
388:
383:
378:
370:
365:
360:
354:
349:
348:
345:
344:
341:
340:
335:
330:
325:
320:
315:
310:
305:
297:
296:
295:
290:
285:
275:
273:Decision trees
270:
264:
250:classification
240:
239:
238:
235:
234:
231:
230:
225:
220:
215:
210:
205:
200:
195:
190:
185:
180:
175:
170:
165:
160:
155:
150:
145:
143:Classification
139:
136:
135:
132:
131:
128:
127:
122:
117:
112:
107:
102:
100:Batch learning
97:
92:
87:
82:
77:
72:
67:
61:
58:
57:
54:
53:
42:
41:
26:
24:
14:
13:
10:
9:
6:
4:
3:
2:
1908:
1897:
1894:
1892:
1889:
1888:
1886:
1872:
1868:
1863:
1858:
1854:
1850:
1846:
1839:
1836:
1831:
1824:
1821:
1816:
1812:
1808:
1804:
1799:
1794:
1790:
1786:
1779:
1776:
1770:
1765:
1758:
1755:
1750:
1743:
1740:
1735:
1728:
1726:
1724:
1722:
1720:
1716:
1711:
1707:
1700:
1697:
1690:
1682:
1678:
1672:
1669:
1662:
1658:
1655:
1653:
1650:
1648:
1645:
1643:
1640:
1638:
1635:
1634:
1630:
1628:
1626:
1622:
1614:
1612:
1586:
1578:
1576:
1574:
1570:
1563:
1561:
1559:
1552:
1550:
1547:
1540:
1538:
1535:
1528:
1526:
1524:
1519:
1512:
1510:
1507:
1500:
1498:
1496:
1489:
1487:
1484:
1475:
1473:
1471:
1467:
1463:
1459:
1455:
1451:
1443:
1441:
1438:
1434:
1430:
1419:
1415:
1410:
1402:
1400:
1397:
1375:
1367:
1363:
1352:
1349:
1346:
1340:
1337:
1328:
1322:
1313:
1303:
1302:
1301:
1265:
1259:
1255:
1249:
1244:
1240:
1233:
1229:
1222:
1218:
1209:
1208:
1198:
1197:
1195:
1175:
1174:
1172:
1146:
1140:
1134:
1133:
1131:
1130:
1129:
1122:
1120:
1116:
1112:
1104:
1102:
1100:
1096:
1088:
1083:
1080:
1077:
1076:
1075:
1066:
1063:
1055:
1045:
1041:
1035:
1034:
1029:This section
1027:
1023:
1018:
1017:
1011:
1009:
1007:
1002:
1000:
996:
991:
989:
985:
981:
977:
973:
961:
956:
954:
949:
947:
942:
941:
939:
938:
931:
928:
924:
921:
920:
919:
916:
914:
911:
910:
904:
903:
896:
893:
891:
888:
886:
883:
881:
878:
876:
873:
871:
868:
866:
863:
862:
856:
855:
848:
845:
843:
840:
838:
835:
833:
830:
828:
825:
823:
820:
818:
815:
813:
810:
809:
803:
802:
795:
792:
790:
787:
785:
782:
780:
777:
776:
770:
769:
762:
759:
757:
754:
752:
751:Crowdsourcing
749:
747:
744:
743:
737:
736:
727:
724:
723:
722:
719:
717:
714:
712:
709:
707:
704:
703:
700:
695:
694:
686:
683:
681:
680:Memtransistor
678:
676:
673:
671:
668:
664:
661:
660:
659:
656:
654:
651:
647:
644:
642:
639:
637:
634:
632:
629:
628:
627:
624:
622:
619:
617:
614:
612:
609:
605:
602:
601:
600:
597:
593:
590:
588:
585:
583:
580:
578:
575:
574:
573:
570:
568:
565:
563:
562:Deep learning
560:
558:
555:
554:
551:
546:
545:
538:
535:
533:
530:
528:
526:
522:
520:
517:
516:
513:
508:
507:
498:
497:Hidden Markov
495:
493:
490:
488:
485:
484:
483:
480:
479:
476:
471:
470:
463:
460:
458:
455:
453:
450:
448:
445:
443:
440:
438:
435:
433:
430:
428:
425:
423:
420:
419:
416:
411:
410:
403:
400:
398:
395:
393:
389:
387:
384:
382:
379:
377:
375:
371:
369:
366:
364:
361:
359:
356:
355:
352:
347:
346:
339:
336:
334:
331:
329:
326:
324:
321:
319:
316:
314:
311:
309:
306:
304:
302:
298:
294:
293:Random forest
291:
289:
286:
284:
281:
280:
279:
276:
274:
271:
269:
266:
265:
258:
257:
252:
251:
243:
237:
236:
229:
226:
224:
221:
219:
216:
214:
211:
209:
206:
204:
201:
199:
196:
194:
191:
189:
186:
184:
181:
179:
178:Data cleaning
176:
174:
171:
169:
166:
164:
161:
159:
156:
154:
151:
149:
146:
144:
141:
140:
134:
133:
126:
123:
121:
118:
116:
113:
111:
108:
106:
103:
101:
98:
96:
93:
91:
90:Meta-learning
88:
86:
83:
81:
78:
76:
73:
71:
68:
66:
63:
62:
56:
55:
52:
47:
43:
39:
38:
33:
19:
1852:
1848:
1838:
1823:
1788:
1784:
1778:
1757:
1748:
1742:
1733:
1709:
1699:
1680:
1671:
1618:
1582:
1567:
1556:
1544:
1532:
1516:
1504:
1493:
1479:
1447:
1439:
1432:
1428:
1417:
1413:
1408:
1406:
1393:
1291:
1257:
1253:
1247:
1242:
1238:
1231:
1227:
1220:
1216:
1123:
1119:scoring rule
1114:
1110:
1108:
1105:One-vs.-rest
1098:
1094:
1092:
1073:
1058:
1049:
1038:Please help
1033:verification
1030:
1003:
992:
983:
979:
969:
837:PAC learning
524:
373:
368:Hierarchical
300:
254:
248:
1855:: 820â836.
1791:: 310â321.
1736:. Springer.
1518:Naive Bayes
1513:Naive Bayes
1462:naive Bayes
1409:one-vs.-one
1403:One-vs.-one
1196:Procedure:
1111:one-vs.-all
1095:one vs rest
721:Multi-agent
658:Transformer
557:Autoencoder
313:Naive Bayes
51:data mining
1885:Categories
1862:2404.16958
1798:1609.00085
1769:2305.03238
1691:References
1615:Evaluation
1278:to obtain
1203:in {1, âŚ,
1099:one vs one
1052:April 2021
706:Q-learning
604:Restricted
402:Mean shift
351:Clustering
328:Perceptron
256:regression
158:Clustering
153:Regression
1396:heuristic
1350:…
1341:∈
1317:^
1263:otherwise
1199:For each
1187:â {1, âŚ,
1113:, OvA or
865:ECML PKDD
847:VC theory
794:ROC curve
726:Self-play
646:DeepDream
487:Bayes net
278:Ensembles
59:Paradigms
1815:12510650
1631:See also
1625:macro F1
1621:Accuracy
1435:â 1) / 2
1420:â 1) / 2
1173:Output:
1158:â {1, âŚ
1141:samples
1132:Inputs:
288:Boosting
137:Problems
1407:In the
1147:labels
870:NeurIPS
687:(ECRAM)
641:AlexNet
283:Bagging
1813:
1266:Apply
1214:where
1151:where
999:binary
663:Vision
519:RANSAC
397:OPTICS
392:DBSCAN
376:-means
183:AutoML
1857:arXiv
1811:S2CID
1793:arXiv
1764:arXiv
1663:Notes
1251:and
885:IJCAI
711:SARSA
670:Mamba
636:LeNet
631:U-Net
457:t-SNE
381:Fuzzy
358:BIRCH
1573:tree
1468:and
1183:for
1097:and
974:and
895:JMLR
880:ICLR
875:ICML
761:RLHF
577:LSTM
363:CURE
49:and
1867:doi
1803:doi
1789:207
1675:In
1623:or
1607:, y
1333:max
1329:arg
1270:to
1261:= 0
1236:if
1042:by
982:or
970:In
621:SOM
611:GAN
587:ESN
582:GRU
527:-NN
462:SDL
452:PGD
447:PCA
442:NMF
437:LDA
432:ICA
427:CCA
303:-NN
1887::
1865:.
1853:12
1851:.
1847:.
1809:.
1801:.
1787:.
1718:^
1708:.
1627:.
1525:.
1464:,
1460:,
1456:,
1452:,
1274:,
1246:=
1225:=
1207:}
978:,
890:ML
1873:.
1869::
1859::
1832:.
1817:.
1805::
1795::
1772:.
1766::
1751:.
1712:.
1609:t
1605:t
1601:t
1597:t
1593:t
1589:t
1433:K
1431:(
1429:K
1424:K
1418:K
1416:(
1414:K
1379:)
1376:x
1373:(
1368:k
1364:f
1356:}
1353:K
1347:1
1344:{
1338:k
1323:=
1314:y
1298:k
1294:x
1282:k
1280:f
1276:z
1272:X
1268:L
1258:i
1254:z
1248:k
1243:i
1239:y
1232:i
1228:y
1221:i
1217:z
1212:z
1205:K
1201:k
1191:}
1189:K
1185:k
1180:k
1178:f
1166:i
1164:X
1160:K
1155:i
1153:y
1149:y
1143:X
1136:L
1126:L
1065:)
1059:(
1054:)
1050:(
1036:.
959:e
952:t
945:v
525:k
374:k
301:k
259:)
247:(
34:.
20:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.