File size: 86,597 Bytes
6fa4bc9 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 | {
"paper_id": "P01-1049",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T09:29:40.314600Z"
},
"title": "Building Semantic Perceptron Net for Topic Spotting",
"authors": [
{
"first": "Jimin",
"middle": [],
"last": "Liu",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "National University of Singapore",
"location": {
"postCode": "117543",
"country": "SINGAPORE"
}
},
"email": ""
},
{
"first": "Tat-Seng",
"middle": [],
"last": "Chua",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "National University of Singapore",
"location": {
"postCode": "117543",
"country": "SINGAPORE"
}
},
"email": "chuats@comp.nus.edu.sg"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "This paper presents an approach to automatically build a semantic perceptron net (SPN) for topic spotting. It uses context at the lower layer to select the exact meaning of key words, and employs a combination of context, co-occurrence statistics and thesaurus to group the distributed but semantically related words within a topic to form basic semantic nodes. The semantic nodes are then used to infer the topic within an input document. Experiments on Reuters 21578 data set demonstrate that SPN is able to capture the semantics of topics, and it performs well on topic spotting task.",
"pdf_parse": {
"paper_id": "P01-1049",
"_pdf_hash": "",
"abstract": [
{
"text": "This paper presents an approach to automatically build a semantic perceptron net (SPN) for topic spotting. It uses context at the lower layer to select the exact meaning of key words, and employs a combination of context, co-occurrence statistics and thesaurus to group the distributed but semantically related words within a topic to form basic semantic nodes. The semantic nodes are then used to infer the topic within an input document. Experiments on Reuters 21578 data set demonstrate that SPN is able to capture the semantics of topics, and it performs well on topic spotting task.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "Topic spotting is the problem of identifying the presence of a predefined topic in a text document. More formally, given a set of n topics together with a collection of documents, the task is to determine for each document the probability that one or more topics is present in the document. Topic spotting may be used to automatically assign subject codes to newswire stories, filter electronic emails and online news, and pre-screen document in information retrieval and information extraction applications.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "Topic spotting, and its related problem of text categorization, has been a hot area of research for over a decade. A large number of techniques have been proposed to tackle the problem, including: regression model, nearest neighbor classification, Bayesian probabilistic model, decision tree, inductive rule learning, neural network, on-line learning, and, support vector machine (Yang & Liu, 1999; Tzeras & Hartmann, 1993) . Most of these methods are word-based and consider only the relationships between the features and topics, but not the relationships among features.",
"cite_spans": [
{
"start": 380,
"end": 398,
"text": "(Yang & Liu, 1999;",
"ref_id": "BIBREF27"
},
{
"start": 399,
"end": 423,
"text": "Tzeras & Hartmann, 1993)",
"ref_id": "BIBREF24"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "It is well known that the performance of the word-based methods is greatly affected by the lack of linguistic understanding, and, in particular, the inability to handle synonymy and polysemy. A number of simple linguistic techniques has been developed to alleviate such problems, ranging from the use of stemming, lexical chain and thesaurus (Jing & Tzoukermann, 1999; Green, 1999) , to word-sense disambiguation (Chen & Chang, 1998; Leacock et al, 1998; Ide & Veronis, 1998) and context (Cohen & Singer, 1999; Jing & Tzoukermann, 1999) .",
"cite_spans": [
{
"start": 342,
"end": 368,
"text": "(Jing & Tzoukermann, 1999;",
"ref_id": "BIBREF11"
},
{
"start": 369,
"end": 381,
"text": "Green, 1999)",
"ref_id": "BIBREF8"
},
{
"start": 413,
"end": 433,
"text": "(Chen & Chang, 1998;",
"ref_id": null
},
{
"start": 434,
"end": 454,
"text": "Leacock et al, 1998;",
"ref_id": null
},
{
"start": 455,
"end": 475,
"text": "Ide & Veronis, 1998)",
"ref_id": "BIBREF10"
},
{
"start": 488,
"end": 510,
"text": "(Cohen & Singer, 1999;",
"ref_id": "BIBREF5"
},
{
"start": 511,
"end": 536,
"text": "Jing & Tzoukermann, 1999)",
"ref_id": "BIBREF11"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "The connectionist approach has been widely used to extract knowledge in a wide range of information processing tasks including natural language processing, information retrieval and image understanding (Anderson, 1983; Lee & Dubin, 1999; Sarkas & Boyer, 1995; Wang & Terman, 1995) . Because the connectionist approach closely resembling human cognition process in text processing, it seems natural to adopt this approach, in conjunction with linguistic analysis, to perform topic spotting. However, there have been few attempts in this direction. This is mainly because of difficulties in automatically constructing the semantic networks for the topics.",
"cite_spans": [
{
"start": 202,
"end": 218,
"text": "(Anderson, 1983;",
"ref_id": "BIBREF0"
},
{
"start": 219,
"end": 237,
"text": "Lee & Dubin, 1999;",
"ref_id": "BIBREF16"
},
{
"start": 238,
"end": 259,
"text": "Sarkas & Boyer, 1995;",
"ref_id": "BIBREF22"
},
{
"start": 260,
"end": 280,
"text": "Wang & Terman, 1995)",
"ref_id": "BIBREF26"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "In this paper, we propose an approach to automatically build a semantic perceptron net (SPN) for topic spotting. The SPN is a connectionist model with hierarchical structure. It uses a combination of context, co-occurrence statistics and thesaurus to group the distributed but semantically related words to form basic semantic nodes. The semantic nodes are then used to identify the topic. This paper discusses the design, implementation and testing of an SPN for topic spotting.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "The paper is organized as follows. Section 2 discusses the topic representation, which is the prototype structure for SPN. Sections 3 & 4 respectively discuss our approach to extract the semantic correlations between words, and build semantic groups and topic tree. Section 5 describes the building and training of SPN, while Section 6 presents the experiment results. Finally, Section 7 concludes the paper.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "The frame of Minsky (1975) is a well-known knowledge representation technique. A frame represents a high-level concept as a collection of slots, where each slot describes one aspect of the concept. The situation is similar in topic spotting. For example, the topic \" water \" may have many aspects (or sub-topics). One sub-topic may be about \"water supply \", while the other is about \"water and environment protection\", and so on. These sub-topics may have some common attributes, such as the word \"water\", and each subtopic may be further sub-divided into finer subtopics, etc.",
"cite_spans": [
{
"start": 13,
"end": 26,
"text": "Minsky (1975)",
"ref_id": "BIBREF19"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Topic Representation",
"sec_num": "2."
},
{
"text": "The above points to a hierarchical t opic representation, which corresponds to the hierarchy of document classes (Figure 1 ). In the model, the contents of the topics and sub-topics (shown as circles) are modeled by a set of attributes, which is simply a group of semantically related words (shown as solid elliptical shaped bags or rectangles). The context (shown as dotted ellipses) is used to identify the exact meaning of a word. Hofmann (1998) presented a word occurrence based cluster abstraction model that learns a hierarchical topic representation. However, the method is not suitable when the set of training examples is sparse. To avoid the problem of automatically constructing the hierarchical model, Tong et al (1987) required the users to supply the model, which is used as queries in the system. Most automated methods, however, avoided this problem by modeling the topic as a feature vector, rule set, or instantiated example (Yang & Liu, 1999) . These methods typically treat each word feature as independent, and seldom consider linguistic factors such as the context or lexical chain relations among the features. As a result, these methods are not good at discriminating a large number of documents that typically lie near the boundary of two or more topics.",
"cite_spans": [
{
"start": 434,
"end": 448,
"text": "Hofmann (1998)",
"ref_id": "BIBREF9"
},
{
"start": 714,
"end": 731,
"text": "Tong et al (1987)",
"ref_id": "BIBREF23"
},
{
"start": 943,
"end": 961,
"text": "(Yang & Liu, 1999)",
"ref_id": "BIBREF27"
}
],
"ref_spans": [
{
"start": 113,
"end": 122,
"text": "(Figure 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Topic Representation",
"sec_num": "2."
},
{
"text": "In order to facilitate the automatic extraction and modeling of the semantic aspects of topics, we adopt a compromise approach. We model the topic as a tree of concepts as shown in Figure 1 . However, we consider only one level of hierarchy built from groups of semantically related words. These semantic groups may not correspond strictly to sub-topics within the domain. Figure 2 shows an example of an automatically constructed topic tree on \"water \". Figure 2 , node \" a\" contains the common feature set of the topic; while nodes \" b\", \"c\" and \"d\" are related to sub-topics on \"water supply\", \"rainfall\", and \"water and environment protection\" respectively. Node \" e\" is the context of the word \"plant\", and node \"f\" is the context of the word \"bank\". Here we use training to automatically resolve the corresponding relationship between a node and an attribute, and the context word to be used to select the exact meaning of a word. From this representation, we observe that: a) Nodes \"c\" and \"d\" are closely related and may not be fully separable. In fact, it is sometimes difficult even for human experts to decide how to divide them into separate topics.",
"cite_spans": [],
"ref_spans": [
{
"start": 181,
"end": 189,
"text": "Figure 1",
"ref_id": null
},
{
"start": 373,
"end": 381,
"text": "Figure 2",
"ref_id": "FIGREF1"
},
{
"start": 455,
"end": 463,
"text": "Figure 2",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "Topic Representation",
"sec_num": "2."
},
{
"text": "b) The same word, such as \"water \", may appear in both the context node and the basic semantic node. c) Some words use context to resolve their meanings, while many do not need context.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Topic Representation",
"sec_num": "2."
},
{
"text": "Although there exists many methods to derive the semantic correlations between words (Lee, 1999; Lin, 1998; Karov & Edelman, 1998; Resnik, 1995; Dagan et al, 1995) , we adopt a relatively simple and yet practical and effective approach to derive three topic -oriented semantic correlations: thesaurus-based, co-occurrence-based and contextbased correlation.",
"cite_spans": [
{
"start": 85,
"end": 96,
"text": "(Lee, 1999;",
"ref_id": "BIBREF15"
},
{
"start": 97,
"end": 107,
"text": "Lin, 1998;",
"ref_id": "BIBREF17"
},
{
"start": 108,
"end": 130,
"text": "Karov & Edelman, 1998;",
"ref_id": "BIBREF12"
},
{
"start": 131,
"end": 144,
"text": "Resnik, 1995;",
"ref_id": "BIBREF21"
},
{
"start": 145,
"end": 163,
"text": "Dagan et al, 1995)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Correlations",
"sec_num": "3."
},
{
"text": "WordNet is an electronic thesaurus popularly used in many researches on lexical semantic acquisition, and word sense disambiguation (Green, 1999; Leacock et al, 1998) . In WordNet, the sense of a word is represented by a list of synonyms (synset), and the lexical information is represented in the form of a semantic network. However, it is well known that the granularity of semantic meanings of words in WordNet is often too fine for practical use. We thus need to enlarge the semantic granularity of words in practical applications. For example, given a topic on \"children education\", it is highly likely that the word \" child\" will be a key term. However, the concept \" child\" can be expressed in many semantically relate d terms, such as \"boy\", \" girl \", \"kid\", \" child\", \" youngster\", etc. In this case, it might not be necessary to distinguish the different meaning among these words, nor the different senses within each word. It is, however, important to group all these words into a large synset {child, boy, girl, kid, youngster}, and use the synset to model the dominant but more general meaning of these words in the context.",
"cite_spans": [
{
"start": 132,
"end": 145,
"text": "(Green, 1999;",
"ref_id": "BIBREF8"
},
{
"start": 146,
"end": 166,
"text": "Leacock et al, 1998)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus based correlation",
"sec_num": "3.1"
},
{
"text": "In general, it is reasonable and often useful to group lexically related words together to represent a more general concept. Here, two words are considered to be lexically related if they are related to by the \"is_a\", \"part_of\", \"member_of\", or \"antonym\" relations, or if they belong to the same synset. Figure 3 lists the lexical relations that we considered, and the examples.",
"cite_spans": [],
"ref_spans": [
{
"start": 304,
"end": 312,
"text": "Figure 3",
"ref_id": "FIGREF2"
}
],
"eq_spans": [],
"section": "Thesaurus based correlation",
"sec_num": "3.1"
},
{
"text": "Since in our experiment, there are many antonyms co-occur within the topic, we also group antonyms together to identify a topic. Moreover, if a word had two senses of, say, sense-1 and sense-2. And if there are two separate words that a re lexically related to this word by sense-1 and sense-2 respectively, we simply group these words together and do not attempt to distinguish the two different senses. The reason is because if a word is so important to be chosen as the keyword of a topic, then it should only have one dominant meaning in that topic. The idea that a keyword should have only one dominant meaning in a topic is also suggested in Church & Yarowsky (1992) . Based on the above discussion, we compute the thesaurus-based correlation between the two terms t 1 and t 2 , in topic T i , as: 1 (t1 and t2 are in the same synset, or t1=t2) 0.8 (t1 and t2 have \"antonym\" relation) 0..5 (t1 and t 2 have relations of \"is_a\", \"part_of\", or \"member_of\") 0 (others)",
"cite_spans": [
{
"start": 648,
"end": 672,
"text": "Church & Yarowsky (1992)",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus based correlation",
"sec_num": "3.1"
},
{
"text": "= ) , ( 2 1 ) ( t t R i L",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus based correlation",
"sec_num": "3.1"
},
{
"text": "Co-occurrence relationship is like the global context of words. Using co-occurrence statistics, Veling & van der Weerd (1999) was able to find many interesting conceptual groups in the Reuters-2178 text corpus. Examples of the conceptual groups found include: { water, rainfall, dry}, {bomb, injured, explosion, injuries}, and {cola, PEP, Pepsi, Pespi-cola, Pepsico}. These groups are meaningful, and are able to capture the important concepts within the corpus.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Co-occurrence based correlation",
"sec_num": "3.2"
},
{
"text": "Since in general, high co-occurrence words are likely to be used together to represent (or describe) a certain concept, i t is reasonable to group them together to form a large semantic node. Thus for topic T i , the co-occurrence-based correlation of two terms, t 1 and t 2 , is computed as:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Co-occurrence based correlation",
"sec_num": "3.2"
},
{
"text": ") ( / ) ( ) , ( 2 1 ) ( 2 1 ) ( 2 1 ) ( t t df t t df t t R i i i co \u2228 \u2227 = (2) where ) ( 2 1 ) ( t t df i \u2227 ( ) ( 2 1 ) ( t t df i \u2228",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Co-occurrence based correlation",
"sec_num": "3.2"
},
{
"text": ") is the fraction of documents in T i that contains t 1 and (or) t 2 .",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Co-occurrence based correlation",
"sec_num": "3.2"
},
{
"text": "Broadly speaking, there are three kinds of context: domain, topic and local contexts (Ide & Vernois, 1998) . Domain context requires extensive knowledge of domain and i s not considered in this paper. Topic context can be modeled approximately using the co-occurrence 1relationships between the words in the topic. In this section, we will define the local context explicitly. The local context of a word t is often defined as the set of non-trivial words near t. Here a word wd is said to be near t if their word distance is less than a given threshold, which is set to be 5 in our experiment.",
"cite_spans": [
{
"start": 85,
"end": 106,
"text": "(Ide & Vernois, 1998)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "We represent the local context of term t j in topic T i by a context vector cv (i) (t j ). To derive cv (i) ",
"cite_spans": [
{
"start": 104,
"end": 107,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "(t j )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": ", we first rank all candidate context words of t i by their density values:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": ") ( / ) ( ) ( ) ( ) ( j i k i j i jk t n wd m = \u03c1 (3) where ) ( ) ( j i t n",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "is the number of occurrence of t j in T i , and",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": ") ( ) ( k i j wd m",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "is the number of occurrences of wd k near t j . We then select from the ranking, the top ten words as the context of t j in T i as:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": ") , ( ),..., , ( ), , {( ) ( ) ( 10 ) ( 10 ) ( 2 ) ( 2 ) ( 1 ) ( 1 ) ( i j i j i j i j i j i j j i wd wd wd t cv \u03c1 \u03c1 \u03c1 = (4)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "When the training sample is sufficiently large, the context vector will have good statistic meanings. Noting again that an important word to a topic should have only one dominant meaning within that topic, and this meaning should be reflected by its context. We can thus draw the conclusion that if two words have a very high context similarity within a topic, it will have a high possibility that they are semantic related. Therefore it is reasonable to group them together to form a larger semantic node. We thus compute the context-based correlation between two term t 1 and t 2 in topic T i as:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "2 / 1 2 ) ( 2 2 / 1 2 ) ( 1 10 1 ) ( ) ( 2 ) ( 1 ) ( ) ( 2 ) ( 1 ) ( 2 1 ) ( ] ) ( [ * ] ) ( [ * * ) , ( ) , ( \u2211 \u2211 \u2211 = = k i k k i k k i k m i k i k m i k i co i c wd wd R t t R \u03c1 \u03c1 \u03c1 \u03c1 (5) where ) , ( max arg ) ( ) ( 2 ) ( 1 ) ( i s i k i co s wd wd R k m =",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "For example, in Reuters 21578 corpus, \"company\" and \"corp\" are context-related words within the topic \"acq\". This is because they have very similar context of \" say, header, acquire, contract\".",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Context based correlation",
"sec_num": "3.3"
},
{
"text": "There are many methods that attempt to construct the conceptual representation of a topic from the original data set (Veling & van der Weerd, 1999; Baker & McCallum, 1998; Pereira et al, 1993) . In this Section, we will describe our semantic -based approach to finding basic semantic g roups and constructing the topic tree. Given a set of training documents, the stages involved in finding the semantic groups for each topic are given below. A) Extract all distinct terms {t 1 , t 2 , ..t n } from the training document set for topic T i . For each term t j , compute its df (i) (t j ) and cv (i) ",
"cite_spans": [
{
"start": 117,
"end": 147,
"text": "(Veling & van der Weerd, 1999;",
"ref_id": "BIBREF25"
},
{
"start": 148,
"end": 171,
"text": "Baker & McCallum, 1998;",
"ref_id": null
},
{
"start": 172,
"end": 192,
"text": "Pereira et al, 1993)",
"ref_id": "BIBREF20"
},
{
"start": 594,
"end": 597,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "(t j )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": ", where df (i) ",
"cite_spans": [
{
"start": 11,
"end": 14,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "(t j )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "is defined as the fraction of documents in T i that contain t j . In other words, df (i) (t j ) gives the conditional probability of t j appearing in T i . B) Derive the semantic group G j using t j as the main keyword. Here we use the semantic correlations defined in Section 3 to derive the semantic relationship between t j and any other term t k . Thus: For each pair (t j ,t k ), k=1,..n, set",
"cite_spans": [
{
"start": 85,
"end": 88,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [
{
"start": 89,
"end": 95,
"text": "(t j )",
"ref_id": null
}
],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "Link(t j ,t k )=1 if ) ( i L R (t j ,t k )>0, or, df (i) (t j )>d 0 and ) (i co R (t j , t k )>d 1 or df (i) (t j )>d 2 and ) (i c R (t j , t k )>d 3 . where d 0 , d 1 , d 2 , d 3 are predefined thresholds.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "For all t k with Link(t j ,t k )=1, we form a semantic group centered around t j denoted by:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "} ,..., , { } ,..., , { 2 1 2 1 n j j j j t t t t t t G j k \u2286 = (6)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "Here t j is the main keyword of node G j and is denoted by main(G j )=t j . C) Calculate the information value inf (i) (G j ) of each basic semantic group. First we compute the information value of each t j :",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "} 1 , 0 max{ * ) ( ) ( inf ) ( ) ( N p t df t ij j i j i \u2212 = (7) where \u2211 = = N k k i j i ij t df t df p 1 ) ( ) ( ) ( ) (",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "and N is the number of topics. Thus 1/N denotes the probability that a term is in any class, and p ij denotes the normalized conditional probability of t j in T i . Only those terms whose normalized conditional probability is higher than 1/N will have a positive information value. The information value of the semantic group G j is simply the summation of information value of its constituent terms weighted by their maximum semantic correlation with t j as:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "\u2211 = = j k k k i i jk j i t w G 1 ) ( ) ( ) ( )] ( inf * [ ) ( inf (8) where )} , ( ), , ( ), , ( max{ ) ( ) ( ) ( ) ( k j i L k j i c k j i co i jk t t R t t R t t R w = D)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "Select the essential semantic groups using the following algorithm: a) Initialize:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "} ,..., , { 1 1 n G G G S \u2190 , \u03a6 \u2190 Groups , b",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": ") Select the semantic group with highest information value: ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": ")) ( (inf max arg ) ( k i S G k G j k \u2208 \u2190 c) Terminate if inf (i) (G j )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "\u03a6 = k G , then } { k G S S \u2212 \u2190 g) If \u03a6 = S",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "then stop; else go to step (b). In the above grouping algorithm, the predefined thresholds d 0 ,d 1 ,d 2 ,d 3 are used to control the size of each group, and d 4 is used to control the number of groups.",
"cite_spans": [],
"ref_spans": [
{
"start": 93,
"end": 107,
"text": "0 ,d 1 ,d 2 ,d",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "The set of basic semantic groups found then forms the sub-topics of a 2-layered topic tree as illustrated in Figure 2 .",
"cite_spans": [],
"ref_spans": [
{
"start": 109,
"end": 117,
"text": "Figure 2",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "Semantic Groups & Topic Tree",
"sec_num": "4."
},
{
"text": "The Combination of local perception and global arbitrator has been applied to solve perception problems (Wang & Terman, 1995; Liu & Shi, 2000) . Here we adopt the same strategy for topic spotting. For each topic, we construct a local perceptron net (LPN), which is designed for a particular topic. We use a global expert (GE) to arbitrate all decisions of LPNs and to model the relationships between topics. Here we discuss the design of both LPN and GE, and their training processes.",
"cite_spans": [
{
"start": 104,
"end": 125,
"text": "(Wang & Terman, 1995;",
"ref_id": "BIBREF26"
},
{
"start": 126,
"end": 142,
"text": "Liu & Shi, 2000)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Building and Training of SPN",
"sec_num": "5."
},
{
"text": "We derive the LPN directly from the topic tree as discussed in Sectio n 2 (see Figure 2) . Each LPN is a multi -layer feed-forward neural network with a typical structure as shown in Figure 4 .",
"cite_spans": [],
"ref_spans": [
{
"start": 79,
"end": 88,
"text": "Figure 2)",
"ref_id": "FIGREF1"
},
{
"start": 183,
"end": 191,
"text": "Figure 4",
"ref_id": null
}
],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "In Figure 4 , x ij represents the feature value of keyword wd ij in the i th semantic group; x ijk 's (where k=1,\u202610) represent the feature values of the context words wd ijk 's of keyword wd ij ; and a ij denotes the meaning of keyword wd ij as determined by its context. A i corresponds to the i th basic semantic node. The weights w i , w ij , and w ijk and biases \u00e8 i and \u00e8 ij are learned from training, and y (i) (x) is the output of the network. \u03b8 (i) ij \u03b8 Figure 4 : The architecture of LPN for topic i Given a document:",
"cite_spans": [
{
"start": 414,
"end": 417,
"text": "(i)",
"ref_id": null
},
{
"start": 454,
"end": 457,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [
{
"start": 3,
"end": 11,
"text": "Figure 4",
"ref_id": null
},
{
"start": 418,
"end": 421,
"text": "(x)",
"ref_id": null
},
{
"start": 463,
"end": 471,
"text": "Figure 4",
"ref_id": null
}
],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "x = {(x ij ,cv ij ) | i=1,2,\u2026m, j=1,\u2026i j }",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "where m is the number of basic semantic nodes, i j is the number of key terms contained in the i th semantic node, and",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "cv ij ={x ij1 ,x ij2 \u2026 ij ijk",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "x } is the context of term x ij . The output y (i) =y (i) (x) is calculated as follows:",
"cite_spans": [
{
"start": 54,
"end": 57,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "\u2211 = = = m i i i i i A w x y y 1 ) ( ) ( ) ( (9) where ] ) * ( exp[ 1 1 * \u2211 \u2212 \u2212 + = \u2208 ij ijk cv x ij ijk ijk ij ij x w x a \u03b8 (10) and ) exp( 1 ) exp( 1 1 1 \u2211 \u2212 + \u2211 \u2212 \u2212 = = = j j i j ij i i j ij i i a w a w A (11)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "Equation (10) expresses the fact that only if a key term is present in the document (i.e. xij > 0), its context needs to be checked.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "For each topic T i , there is a corresponding net y (i) =y (i) (x) and a threshold \u03b8 (i) . The pair of (y (i) ",
"cite_spans": [
{
"start": 59,
"end": 62,
"text": "(i)",
"ref_id": null
},
{
"start": 85,
"end": 88,
"text": "(i)",
"ref_id": null
},
{
"start": 106,
"end": 109,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "(x), \u03b8 (i) )",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "is a local binary classifier for T i such that:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "If y (i) (x)-\u03b8 (i) > 0, then T i is present; otherwise",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "T i is not present in document x. From the procedures employed to building the topic tree, we know that each feature is in fact an evidence to support the occurrence of the topic. This gives us the suggestion that the activation function for each node in the LPN should be a nondecreasing function of the inputs. Thus we impose a weight constraint on the LPN as: w i >0, w ij >0, w ijk >0 (12)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Local Perceptron Net (LPN)",
"sec_num": "5.1"
},
{
"text": "Since there are relations among topics, and LPNs do not have global information, it is inevitable that LPNs will make wrong decisions. In order to overcome this problem, we use a global expert (GE) to arbitrate al local decisions. Figure 5 illustrates the use of global expert to combine the outputs of LPNs. Figure 5 : The architecture of global expert Given a document x, we first use each LPN to make a local decision. We then combine the outputs of LPNs as follows:",
"cite_spans": [],
"ref_spans": [
{
"start": 231,
"end": 239,
"text": "Figure 5",
"ref_id": null
},
{
"start": 309,
"end": 317,
"text": "Figure 5",
"ref_id": null
}
],
"eq_spans": [],
"section": "Global expert (GE)",
"sec_num": "5.2"
},
{
"text": ") ( ) ( i i y \u03b8 \u2212 Y (i) ) ( ) ( j j y \u03b8 \u2212 ij W ) 1 ( ) 1 ( \u03b8 \u2212 y ) (i \u0398",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Global expert (GE)",
"sec_num": "5.2"
},
{
"text": "] ) ( [ ) ( ) ( ) ( ) ( ) ( ) ( ) ( 0 ) ( ) ( i j ij i i i j j j y i j y W y Y \u0398 \u2212 \u2212 \u2211 + \u2212 = > \u2212 \u2260 \u03b8 \u03b8 \u03b8 (13)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Global expert (GE)",
"sec_num": "5.2"
},
{
"text": "where W ij 's are the weights between the global arbitrator i and the j th LPN; and ) ( i \u0398 's are the global bias. From the result of Equation (13), we have:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Global expert (GE)",
"sec_num": "5.2"
},
{
"text": "If Y (i) > 0; then topic T i is present; otherwise T i is not present in document x The use of Equation (13) implies that: a) If a LPN is not activated, i.e., y (i) \u2264 \u03b8 (i) , then its output is not used in the GE. Thus it will not affect the output of other LPN. b) The weight W ij models the relationship or correlation between topic i and j. If W ij > 0, it means that if document x is related to T j , it may also have some contribution ( W ij ) to topic T j . On the other hand, if W ij < 0, it means the two topics are negatively correlated, and a document x will not be related to both T j and T i . The overall structure of SPN is as follows: Figure 6 : Overall structure of SPN",
"cite_spans": [
{
"start": 169,
"end": 172,
"text": "(i)",
"ref_id": null
}
],
"ref_spans": [
{
"start": 650,
"end": 658,
"text": "Figure 6",
"ref_id": null
}
],
"eq_spans": [],
"section": "Global expert (GE)",
"sec_num": "5.2"
},
{
"text": "Input document Local Perception Global Expert x y (i) Y (i)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Global expert (GE)",
"sec_num": "5.2"
},
{
"text": "In order to adopt SPN for topic spotting, we employ the well-known BP algorithm to derive the optimal weights and biases in SPN. The training phase is divided to two stages. The first stage learns a LPN for each topic, while the second stage trains the GE. As the BP algorithm is rather standard, we will discuss only the error functions that we employ to guide the training process.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Training of SPN",
"sec_num": "5.3"
},
{
"text": "In topic spotting, the goal is to achieve both high recall and precision. In particular, we want to allow y(x) to be as large (or as small) as possible in cases when there is no error, or when After the training, we choose the node with the biggest w i value as the common attribute node. Also, we trim the topic representation by removing those words or context words with very small w ij or w ijk values.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Training of SPN",
"sec_num": "5.3"
},
{
"text": "We adopt the following error function to train GE:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Training of SPN",
"sec_num": "5.3"
},
{
"text": "\u2211 \u2211 \u0398 + \u2211 \u0398 = \u0398 = \u2126 \u2208 \u2212 \u2126 \u2208 + \u2212 + n i x i i i x i i i i ij i i x Y x Y W E 1 ] ) ), ( ( ) ), ( ( [ ) , ( \u03b5 \u03b5 (15)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Training of SPN",
"sec_num": "5.3"
},
{
"text": "where + \u2126 i is the set of positive examples of T i .",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The Training of SPN",
"sec_num": "5.3"
},
{
"text": "We employ the ModApte Split version of Reuters-21578 corpus to test our method. In order to ensure that the training is meaningful, we select only those classes that have at least one document in each of the training and test sets. This results in 90 classes in both the training and test sets. After eliminating documents that do not belong to any of these 90 classes, we obtain a training set of 7,770 documents and a test set of 3,019 documents. From the set of training documents, we derive the set of semantic nodes for each topic using the procedures outlined in Section 4. From the training set, we found that the average number of semantic nodes for each topic is 132, and the average number of terms in each node is 2.4. For illustration, Table 1 lists some examples of the semantic nodes that we found. From table 1, we can draw the following general observations. Table 1 : Examples of semantic nodes a) Under the topic \"wheat \", we list four semantic nodes. Node 1 contains the common attribute set of the topic. Node 2 is related to the \"buying and selling of wheat\". Node 3 is related to \"wheat production\"; and node 4 is related to \"the effects of insect on wheat production\". The results show that the automatically extracted basic semantic nodes are meaningful and are able to capture most semantics of a topic. b) Node 1 originally contains two terms \"wheat\"",
"cite_spans": [],
"ref_spans": [
{
"start": 748,
"end": 755,
"text": "Table 1",
"ref_id": null
},
{
"start": 875,
"end": 882,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment and Discussion",
"sec_num": "6."
},
{
"text": "and \"corn\" that belong to the same synset found by looking up WordNet. However, in the training stage, the weight of the word \" corn\" was found to be very small in topic \"wheat\", and hence it was removed from the semantic group. This is similar to the discourse based word sense disambiguation. c) The granularity of information expressed by the semantic nodes may not be the same as what human expert produces. For example, it is possible that a human expert may divide node 2 into two nodes {import} and {export, output}. d) Node 5 contains four words and is formed by analyzing context. Each context vector of the four words has the s ame two components: \"price\" and \" digital number\". Meanwhile, \"rise\" and \"fall\" can also be grouped together by \"antonym\" relation. \"fell\" is actually the past tense of \" fall\". This means that by comparing context, it is possible to group together those words with grammatical variations without performing grammatical analysis. Table 2 summarizes the results of SPN in terms of macro and micro F 1 values (see Yang & Liu (1999) for definitions of the macro and micro F 1 values). For comparison purpose, the Table also lists the results of other TC methods as reported in Yang & Liu (1999) . From the table, it can be seen that the SPN method achieves the best macF1 value. This indicates that the method performs well on classes with a small number of training samples.",
"cite_spans": [
{
"start": 1050,
"end": 1067,
"text": "Yang & Liu (1999)",
"ref_id": "BIBREF27"
},
{
"start": 1212,
"end": 1229,
"text": "Yang & Liu (1999)",
"ref_id": "BIBREF27"
}
],
"ref_spans": [
{
"start": 968,
"end": 975,
"text": "Table 2",
"ref_id": null
},
{
"start": 1148,
"end": 1158,
"text": "Table also",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment and Discussion",
"sec_num": "6."
},
{
"text": "In terms of the micro F 1 measures, SPN outperforms NB, NNet, LSF and KNN, while posting a slightly lower performance than that of SVM. The results are encouraging as they are rather preliminary. We expect the results to improve further by tuning the system ranging from the initial values of various parameters, to the choice of error functions, context, grouping algorithm, and the structures of topic tree and SPN. ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment and Discussion",
"sec_num": "6."
},
{
"text": "In this paper, we proposed an approach to automatically build semanti c perceptron net (SPN) for topic spotting. The SPN is a connectionist model in which context is used to select the exact meaning of a word. By analyzing the context and co-occurrence statistics, and by looking up thesaurus, it is able to group the distributed but semantic related words together to form basic semantic nodes. Experiments on Reuters 21578 show that, to some extent, SPN is able to capture the semantics of topics and it performs well on topic spotting task. It is well known that human expert, whose most prominent characteristic is the ability to understand text documents, have a strong natural ability to spot topics in documents. We are, however, unclear about the nature of human cognition, and with the present state-of-art natural language processing technology, it is still difficult to get an in-depth understanding of a text passage. We believe that our proposed approach provides a promising compromise between full understanding and no understanding.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "7."
},
{
"text": ")), ( (",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "The authors would like to acknowledge the support of the National Science and Technology Board, and the Ministry of Education of Singapore for the provision of a research grant RP3989903 under which this research is carried out.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgment",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "A Spreading Activation Theory of Memory",
"authors": [
{
"first": "J",
"middle": [
"R"
],
"last": "Anderson",
"suffix": ""
}
],
"year": 1983,
"venue": "J. of Verbal Learning & Verbal Behavior",
"volume": "22",
"issue": "3",
"pages": "261--295",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J.R. Anderson (1983). A Spreading Activation Theory of Memory. J. of Verbal Learning & Verbal Behavior, 22(3):261-295.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Distributional Clustering of Words for Text Classification. SIGIR'98",
"authors": [],
"year": null,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Distributional Clustering of Words for Text Classification. SIGIR'98.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Topic Clustering of MRD Senses based on Information Retrieval Technique",
"authors": [],
"year": 1998,
"venue": "",
"volume": "24",
"issue": "",
"pages": "62--95",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J.N. Chen & J.S. Chang (1998). Topic Clustering of MRD Senses based on Information Retrieval Technique. Comp Linguistic, 24(1), 62-95.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "One Sense per Discourse",
"authors": [
{
"first": "G",
"middle": [
"W K"
],
"last": "Church",
"suffix": ""
},
{
"first": "& D",
"middle": [],
"last": "Yarowsky",
"suffix": ""
}
],
"year": 1992,
"venue": "Proc. of 4 th DARPA Speech and Natural Language Workshop",
"volume": "",
"issue": "",
"pages": "233--237",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "G.W.K. Church & D. Yarowsky (1992). One Sense per Discourse. Proc. of 4 th DARPA Speech and Natural Language Workshop. 233-237.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Context-Sensitive Learning Method for T ext Categorization",
"authors": [
{
"first": "W",
"middle": [
"W"
],
"last": "Cohen",
"suffix": ""
},
{
"first": "& Y",
"middle": [],
"last": "Singer",
"suffix": ""
}
],
"year": 1999,
"venue": "ACM Trans. on Information Systems",
"volume": "17",
"issue": "2",
"pages": "141--173",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "W.W. Cohen & Y. Singer (1999). Context- Sensitive Learning Method for T ext Categorization. ACM Trans. on Information Systems, 17(2), 141-173, Apr.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Building Hypertext Links by Computing Semantic Similarity",
"authors": [
{
"first": "S",
"middle": [
"J"
],
"last": "Green",
"suffix": ""
}
],
"year": 1999,
"venue": "IEEE Trans on Knowledge & Data Engr",
"volume": "",
"issue": "5",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "S.J. Green (1999). Building Hypertext Links by Computing Semantic Similarity. IEEE Trans on Knowledge & Data Engr, 11(5).",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Learning and Representing Topic, a Hierarchical Mixture Model for Word Occurrences in Document Databases",
"authors": [
{
"first": "T",
"middle": [],
"last": "Hofmann",
"suffix": ""
}
],
"year": 1998,
"venue": "Workshop on Learning from Text and the Web",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "T. Hofmann (1998). Learning and Representing Topic, a Hierarchical Mixture Model for Word Occurrences in Document Databases. Workshop on Learning from Text and the Web, CMU.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "Introduction to the Special Issue on Word Sense Disambiguation: the State of Art",
"authors": [
{
"first": "&",
"middle": [
"J"
],
"last": "Ide",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Veronis",
"suffix": ""
}
],
"year": 1998,
"venue": "Comp Linguistics",
"volume": "24",
"issue": "1",
"pages": "1--39",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ide & J. Veronis (1998). Introduction to the Special Issue on Word Sense Disambiguation: the State of Art. Comp Linguistics, 24(1), 1-39.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "Information Retrieval based on Context Distance and Morphology",
"authors": [
{
"first": "H",
"middle": [],
"last": "Jing",
"suffix": ""
},
{
"first": "&",
"middle": [
"E"
],
"last": "Tzoukermann",
"suffix": ""
}
],
"year": 1999,
"venue": "SIGIR'99",
"volume": "",
"issue": "",
"pages": "90--96",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "H. Jing & E. Tzoukermann (1999). Information Retrieval based on Context Distance and Morphology. SIGIR'99, 90-96.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Similarity-based Word Sense Disambiguation",
"authors": [
{
"first": "&",
"middle": [
"S"
],
"last": "Karov",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Edelman",
"suffix": ""
}
],
"year": 1998,
"venue": "Computational Linguistics",
"volume": "24",
"issue": "1",
"pages": "41--59",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Karov & S. Edelman (1998). Similarity-based Word Sense Disambiguation, Computational Linguistics, 24(1), 41-59.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Using Corpus Statistics and WordNet for Sense Identification",
"authors": [],
"year": null,
"venue": "Comp. Linguistic",
"volume": "24",
"issue": "1",
"pages": "147--165",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Using Corpus Statistics and WordNet for Sense Identification. Comp. Linguistic, 24(1), 147- 165.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Measure of Distributional Similarity",
"authors": [
{
"first": "L",
"middle": [],
"last": "Lee",
"suffix": ""
}
],
"year": 1999,
"venue": "Proc of 37 th Annual Meeting of ACL",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "L. Lee (1999). Measure of Distributional Similarity. Proc of 37 th Annual Meeting of ACL.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Context-Sensitive Vocabulary Mapping with a Spreading Activation Network",
"authors": [
{
"first": "&",
"middle": [
"D"
],
"last": "Lee",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Dubin",
"suffix": ""
}
],
"year": 1999,
"venue": "SIGIR'99",
"volume": "",
"issue": "",
"pages": "198--205",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lee & D. Dubin (1999). Context-Sensitive Vocabulary Mapping with a Spreading Activation Network. SIGIR'99, 198-205.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "Automatic Retrieval and Clust ering of Similar Words",
"authors": [
{
"first": "D",
"middle": [],
"last": "Lin",
"suffix": ""
}
],
"year": 1998,
"venue": "COLING-ACL'98",
"volume": "",
"issue": "",
"pages": "768--773",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "D. Lin (1998). Automatic Retrieval and Clust ering of Similar Words. In COLING-ACL'98, 768- 773.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "Extracting Prominent Shape by Local Interactions and Global Optimizations",
"authors": [
{
"first": "J",
"middle": [],
"last": "Liu",
"suffix": ""
},
{
"first": "& Z",
"middle": [],
"last": "Shi",
"suffix": ""
}
],
"year": 2000,
"venue": "CVPRIP",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. Liu & Z. Shi (2000). Extracting Prominent Shape by Local Interactions and Global Optimizations. CVPRIP'2000, USA.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "A Framework for Representing Knowledge",
"authors": [
{
"first": "M",
"middle": [
"A"
],
"last": "Minsky",
"suffix": ""
}
],
"year": 1975,
"venue": "",
"volume": "",
"issue": "",
"pages": "211--277",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M.A. Minsky (1975). A Framework for Representing Knowledge. In: Winston P (eds). \"The psychology of computer vision\", McGraw -Hill, New York, 211-277.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Distributional Clustering of English Words. ACL'93",
"authors": [
{
"first": "F",
"middle": [
"C N"
],
"last": "Pereira",
"suffix": ""
},
{
"first": "N",
"middle": [
"Z"
],
"last": "Tishby",
"suffix": ""
},
{
"first": "&",
"middle": [
"L"
],
"last": "Lee",
"suffix": ""
}
],
"year": 1993,
"venue": "",
"volume": "",
"issue": "",
"pages": "183--190",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "F.C.N. Pereira, N.Z. Tishby & L. Lee (1993). Distributional Clustering of English Words. ACL'93, 183-190.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "Using Information Content to Evaluate Semantic Similarity in a Taxonomy",
"authors": [
{
"first": "P",
"middle": [],
"last": "Resnik",
"suffix": ""
}
],
"year": 1995,
"venue": "Proc of IJCAI -95",
"volume": "",
"issue": "",
"pages": "448--453",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "P. Resnik (1995). Using Information Content to Evaluate Semantic Similarity in a Taxonomy. Proc of IJCAI -95, 448-453.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Using Perceptual Inference Network to Manage Vision Processes. Computer Vision & Image Understanding",
"authors": [
{
"first": "S",
"middle": [
"L"
],
"last": "Sarkas & K",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Boyer",
"suffix": ""
}
],
"year": 1995,
"venue": "",
"volume": "62",
"issue": "",
"pages": "27--46",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "S. Sarkas & K.L. Boyer (1995). Using Perceptual Inference Network to Manage Vision Processes. Computer Vision & Image Understanding, 62(1), 27-46.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "Conceptual Information Retrieval using RUBRIC. SIGIR'87",
"authors": [
{
"first": "R",
"middle": [],
"last": "Tong",
"suffix": ""
},
{
"first": "L",
"middle": [],
"last": "Appelbaum",
"suffix": ""
},
{
"first": "V",
"middle": [],
"last": "Askman",
"suffix": ""
},
{
"first": "& J",
"middle": [],
"last": "Cunningham",
"suffix": ""
}
],
"year": 1987,
"venue": "",
"volume": "",
"issue": "",
"pages": "247--253",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "R. Tong, L. Appelbaum, V. Askman & J. Cunningham (1987). Conceptual Information Retrieval using RUBRIC. SIGIR'87, 247-253.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "Automatic Indexing based on Bayesian Inference Networks",
"authors": [
{
"first": "K",
"middle": [],
"last": "Tzeras",
"suffix": ""
},
{
"first": "&",
"middle": [
"S"
],
"last": "Hartmann",
"suffix": ""
}
],
"year": 1993,
"venue": "SIGIR'93",
"volume": "",
"issue": "",
"pages": "22--34",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "K. Tzeras & S. Hartmann (1993). Automatic Indexing based on Bayesian Inference Networks. SIGIR'93, 22-34.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Conceptual Grouping in Word Co-occurrence Networks",
"authors": [
{
"first": "A",
"middle": [],
"last": "Veling & P. Van Der Weerd",
"suffix": ""
}
],
"year": 1999,
"venue": "IJCAI",
"volume": "99",
"issue": "",
"pages": "694--701",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "A. Veling & P. van der Weerd (1999). Conceptual Grouping in Word Co-occurrence Networks. IJCAI 99: 694-701.",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Locally Excitatory Globally Inhibitory Oscillator Networks",
"authors": [
{
"first": "&",
"middle": [
"D"
],
"last": "Wang",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Terman",
"suffix": ""
}
],
"year": 1995,
"venue": "IEEE Trans. Neural Network",
"volume": "6",
"issue": "1",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Wang & D. Terman (1995). Locally Excitatory Globally Inhibitory Oscillator Networks. IEEE Trans. Neural Network. 6(1).",
"links": null
},
"BIBREF27": {
"ref_id": "b27",
"title": "Re-examination of Text Categorization",
"authors": [
{
"first": "&",
"middle": [
"X"
],
"last": "Yang",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Liu",
"suffix": ""
}
],
"year": 1999,
"venue": "SIGIR'99",
"volume": "",
"issue": "",
"pages": "43--49",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yang & X. Liu (1999). Re-examination of Text Categorization. SIGIR'99, 43-49.",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Figure 1. Topic representation"
},
"FIGREF1": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "An example of a topic tree In"
},
"FIGREF2": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Examples of lexical relationship"
},
"FIGREF4": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "and negative training document sets respectively. To achieve this, we adopt a new error function as follows to train the LPN: ensure that the contributions of positive and negative examples are equal."
}
}
}
} |