File size: 82,663 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 |
{
"paper_id": "R11-1032",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T15:05:06.618478Z"
},
"title": "Improving WSD using ISR-WN with Relevant Semantic Trees and SemCor Senses Frequency",
"authors": [
{
"first": "Yoan",
"middle": [],
"last": "Guti\u00e9rrez",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "University of Matanzas",
"location": {
"country": "Cuba"
}
},
"email": "yoan.gutierrez@umcc.cu"
},
{
"first": "Sonia",
"middle": [],
"last": "V\u00e1zquez",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "University of Alicante",
"location": {
"country": "Spain"
}
},
"email": "svazquez@dlsi.ua.es"
},
{
"first": "Andr\u00e9s",
"middle": [],
"last": "Montoyo",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "University of Alicante",
"location": {
"country": "Spain"
}
},
"email": "montoyo@dlsi.ua.es"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "In this paper we concentrate on the resolution of the semantic ambiguity that arises when a given word has several meanings. This specific task is commonly referred to as Word Sense Disambiguation (WSD). We propose a method that obtains the appropriate senses from a multidimensional analysis (using Relevant Semantic Trees). Our method uses different resources WordNet, WordNet Domains, WordNet-Affects and SUMO, combined with senses frequency obtained from SemCor. Our hypothesis is that in WSD it is important to obtain the most frequent senses depending on the type of analyzed context to achieve better results. Finally, in order to evaluate and compare our results, it is presented a comprehensive study and experimental work using the Senseval-2 and Semeval-2 data set, demonstrating that our system obtains better results than other unsupervised systems.",
"pdf_parse": {
"paper_id": "R11-1032",
"_pdf_hash": "",
"abstract": [
{
"text": "In this paper we concentrate on the resolution of the semantic ambiguity that arises when a given word has several meanings. This specific task is commonly referred to as Word Sense Disambiguation (WSD). We propose a method that obtains the appropriate senses from a multidimensional analysis (using Relevant Semantic Trees). Our method uses different resources WordNet, WordNet Domains, WordNet-Affects and SUMO, combined with senses frequency obtained from SemCor. Our hypothesis is that in WSD it is important to obtain the most frequent senses depending on the type of analyzed context to achieve better results. Finally, in order to evaluate and compare our results, it is presented a comprehensive study and experimental work using the Senseval-2 and Semeval-2 data set, demonstrating that our system obtains better results than other unsupervised systems.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "The main goal of knowledge technologies is to provide meaning to the huge quantity of information that our multilingual societies generate day to day. A wide range of advanced techniques are required to progressively automate the knowledge lifecycle. For that, after performing an analysis to large data collections it is necessary to develop different approaches to automatically represent and manage a high-level of meaningful concepts (Montoyo et al., 2005) . Moreover, to be able to create efficient systems of Natural Language Processing (NLP) it is necessary to turn the information extracted from words in plain text into a Concept Level or meaningful word senses. This representation allows to group words with similar meanings according to the context where they appear.",
"cite_spans": [
{
"start": 438,
"end": 460,
"text": "(Montoyo et al., 2005)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "In order to determine the right meanings of words in different contexts WSD systems have been developed. Furthermore, it has been proved that applications such as Machine Translation, Information Extraction, Question Answering, Information Retrieval, Text Classification, and Text Summarization require knowledge about word meanings to obtain better results. So, WSD is considered an essential task for all these applications (Ide and V\u00e9ronis, 1998) . For this reason many research groups are working on WSD, using a wide range of approaches.",
"cite_spans": [
{
"start": 426,
"end": 449,
"text": "(Ide and V\u00e9ronis, 1998)",
"ref_id": "BIBREF16"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Due to the need of evaluating different approaches to show the improvements of NLP tasks it was created the Senseval 1 competition. The first Senseval was in 1998 at Herstmonceux Castle, Succex (England) and after that every three years a new competition takes place. In Senseval, different NLP tasks are defined in order to evaluate systems using the same repositories and corpus. At present, the results obtained in WSD have been going poorer, because the requirements of each corpus are getting more complex. For example, in Senseval-2 (Cotton et al., 2001) the best system obtained a 69% of accuracy in WSD, three years later in Senseval-3 (Snyder and Palmer, 2004) the best results were around 65.2% of accuracy, next in Semeval-1 (Pradhan et al., 2007) a 59.1% of accuracy was obtained and in Semeval-2 (Agirre et al., 2010) was 55.5%.",
"cite_spans": [
{
"start": 539,
"end": 560,
"text": "(Cotton et al., 2001)",
"ref_id": "BIBREF19"
},
{
"start": 644,
"end": 669,
"text": "(Snyder and Palmer, 2004)",
"ref_id": "BIBREF5"
},
{
"start": 726,
"end": 758,
"text": "Semeval-1 (Pradhan et al., 2007)",
"ref_id": null
},
{
"start": 809,
"end": 830,
"text": "(Agirre et al., 2010)",
"ref_id": "BIBREF10"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Due to the fact that the results are still very low in accuracy we want to go deeply in the resolution of semantic ambiguity. Firstly, we have studied the behavior of the baseline Most Frequent Sense (MFS) in each competition. This baseline has been placed among the top places of the rank; for example, in Senseval-2 a system applying this baseline could have been located on the 2 nd place with a 64.58% of accuracy (Preiss, 2006) . In Senseval-3 Denys Yuret of Koc University computed a 60.9% and for the same competition Bart Decadt of University of Antwerp provided a baseline of 62.4%, these results could have been located on 7 th and 5 th positions respectively (Snyder and Palmer, 2004) . In Semeval-1 the baseline was positioned on 9 th place of fourteen systems and for the Semeval-2 competition the MFS baseline was located on 6 th place. As we can see, this probabilistic procedure can obtain effective results on WSD task, but notice that it does not take into account context information.",
"cite_spans": [
{
"start": 418,
"end": 432,
"text": "(Preiss, 2006)",
"ref_id": "BIBREF14"
},
{
"start": 670,
"end": 695,
"text": "(Snyder and Palmer, 2004)",
"ref_id": "BIBREF5"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Taking into account these facts our hypothesis is that for WSD it is important to obtain the most frequent senses combined with contextual information.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "After these considerations, a new question arises: How will we be able to develop a procedure that uses the sense frequencies combined with a technique that takes into account the context information and improves the MFS results?",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "With the aim to answer this question and to demonstrate our hypothesis we present the following contributions:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "\uf0b7 A method that combines MFS with a multidimensional analysis of the context. It uses several semantic resources combined with Relevant Semantic Trees. \uf0b7 An analysis of how the MFS influences on the Relevant Semantic Trees method. \uf0b7 An analysis of the behavior of Relevant Semantic Trees and Most Frequent Senses in each one of the semantic dimensions. \uf0b7 A voting process between MFS and the results of different semantic dimensions. \uf0b7 An exhaustive evaluation of the proposal. \uf0b7 A comparison between our results and the systems in the Senseval-2 and Semeval-2 competitions. In Section 2 we show some related works. Our approach is described in Section 3. The evaluations and analysis are provided in Section 4. Finally, we conclude in Section 5 adding further works. Among all of these resources, ISR-WN has the highest quantity of semantic dimensions aligned, so it is a suitable resource to run our proposal. Next, we present a brief description of ISR-WN.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Integration of Semantic Resources based on WordNet (ISR-WN) (Guti\u00e9rrez et al., 2010b ) is a new resource that allows the integration of several semantic resources mapped to WN. In ISR-WN, WordNet is used as a core to link several resources such as: SUMO (Niles, 2001) , WordNet Domains (WND) (Magnini and Cavaglia, 2000) and WordNet Affect (WNA) (Strapparava and Valitutti, 2004) . As (Guti\u00e9rrez et al., 2010a) describe, the integrator resource provides a software capable to navigate inside the semantic network. In order to apply the multidimensionality that this resource provides, we have analyzed related NLP approaches that take into account semantic dimensionality. Addressed to context analysis we have studied (Magnini et al., 2008) , (V\u00e1zquez et al., 2004) and (Buscaldi et al., 2005) . In these works WSD is performed using the WND resource (domain dimension). (Zouaq et al., 2009) , (Villarejo et al., 2005) among others, conducted a semantic analysis using SUMO ontology (category dimension), and the Relevant Semantic Trees (RST) (Guti\u00e9rrez et al., 2010a) apply several dimensions at once.",
"cite_spans": [
{
"start": 60,
"end": 84,
"text": "(Guti\u00e9rrez et al., 2010b",
"ref_id": "BIBREF22"
},
{
"start": 254,
"end": 267,
"text": "(Niles, 2001)",
"ref_id": "BIBREF12"
},
{
"start": 292,
"end": 320,
"text": "(Magnini and Cavaglia, 2000)",
"ref_id": "BIBREF3"
},
{
"start": 346,
"end": 379,
"text": "(Strapparava and Valitutti, 2004)",
"ref_id": "BIBREF7"
},
{
"start": 385,
"end": 410,
"text": "(Guti\u00e9rrez et al., 2010a)",
"ref_id": "BIBREF21"
},
{
"start": 719,
"end": 741,
"text": "(Magnini et al., 2008)",
"ref_id": "BIBREF4"
},
{
"start": 744,
"end": 766,
"text": "(V\u00e1zquez et al., 2004)",
"ref_id": "BIBREF20"
},
{
"start": 771,
"end": 794,
"text": "(Buscaldi et al., 2005)",
"ref_id": "BIBREF8"
},
{
"start": 872,
"end": 892,
"text": "(Zouaq et al., 2009)",
"ref_id": "BIBREF2"
},
{
"start": 895,
"end": 919,
"text": "(Villarejo et al., 2005)",
"ref_id": "BIBREF15"
},
{
"start": 1044,
"end": 1069,
"text": "(Guti\u00e9rrez et al., 2010a)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Integration of Semantic Resources based on WordNet (ISR-WN)",
"sec_num": "2.1"
},
{
"text": "Next, we present the RST method which is able to work with different resources based on WordNet.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Integration of Semantic Resources based on WordNet (ISR-WN)",
"sec_num": "2.1"
},
{
"text": "The RST method is able to find the correct senses of each word using Relevant Semantic Trees from different resources. This approach can be used with many resources mapped to WN as we have mentioned above.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Relevant Semantic Trees (RST)",
"sec_num": "2.2"
},
{
"text": "In order to measure the association between concepts according to a multidimensional perspective in each sentence, RST uses an Association Ratio (AR) modification based on the proposal presented by V\u00e1zquez et al. (2004) .",
"cite_spans": [
{
"start": 198,
"end": 219,
"text": "V\u00e1zquez et al. (2004)",
"ref_id": "BIBREF20"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Relevant Semantic Trees (RST)",
"sec_num": "2.2"
},
{
"text": "We propose an unsupervised knowledge-based method that uses the original RST technique including senses frequency of SemCor 4 corpus and using a voting process to find the right senses. The voting process involves MFS (Most Frequent Sense), RST over WND, WNA, WN taxonomy and SUMO. Adding this new information we are able to improve the previous results obtained by the original RST and we also improve the MFS results in Semeval-2 competition. Specifically, we provide a sort of supervised aid (i.e. MFS) to the RST method of Guti\u00e9rrez et al.(2010a) Step 3. Voting process to obtain the final senses. Next, we present how these phases have been developed.",
"cite_spans": [
{
"start": 527,
"end": 550,
"text": "Guti\u00e9rrez et al.(2010a)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "WSD Method",
"sec_num": "3"
},
{
"text": "In this section, we describe how we have used a fragment of the original RST method with the aim to obtain Relevant Semantic Trees from the sentences. Equation 1 is used to measure and obtain the values of Relevant Concepts:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "(1) Where 2Where C is a concept; s is a sentence or a set of words (w); s i is the i-th word (w) of the sentence s; P(C, w) is joint probability distribution; and P(C) is marginal probability.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "The first stage is to Pre-process the sentence to obtain all lemmas. For instance, in the sentence \"But it is unfair to dump on teachers as distinct from the educational establishment.\" the lemmas are: [unfair, dump, teacher, distinct, educational, establishment Next, each lemma is searched through ISR-WN resource and it is correlated with concepts of WND (the dimension used in this example). Table 1 shows the results after applying Equation 1 over the sentence.",
"cite_spans": [
{
"start": 202,
"end": 262,
"text": "[unfair, dump, teacher, distinct, educational, establishment",
"ref_id": null
}
],
"ref_spans": [
{
"start": 396,
"end": 403,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "After obtaining the Initial Concept Vector of Domains we apply the Equation 3 in order to build the Relevant Semantic Tree related to the sentence.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "; 3Where:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "; 4Where AR(PC, s) represents the AR value of PC related to the sentence s;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "is the AR value calculated with Equation 1 in case of Child Concept (ChC) was included in the Initial Vector, otherwise is calculated with the Equation 3; ND is a Normalized Distance; IC is the Initial Concept from we have to add the ancestors; PC is Parent Concept; TD is Depth of the hierarchic tree of the resource to use; and MP is Minimal Path.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "Applying the Equation 3, the algorithm to decide which parent concept will be added to the vector is shown here: Figure 1 shows. As we can see, the Relevant Semantic Tree of domains in Figure 1 has associated a color intensity related to the AR value of each domain. The more intense the color is the more related AR is. (Magnini and Cavaglia, 2000) . Moreover, after conducting several experiments we have confirmed that it introduced errors.",
"cite_spans": [
{
"start": 321,
"end": 349,
"text": "(Magnini and Cavaglia, 2000)",
"ref_id": "BIBREF3"
}
],
"ref_spans": [
{
"start": 113,
"end": 121,
"text": "Figure 1",
"ref_id": null
},
{
"start": 185,
"end": 193,
"text": "Figure 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "EQUATION",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [
{
"start": 0,
"end": 8,
"text": "EQUATION",
"ref_id": "EQREF",
"raw_str": "if (",
"eq_num": "> 0"
}
],
"section": "Obtaining the Relevant Semantic Trees",
"sec_num": "3.1"
},
{
"text": "To select the correct senses, three steps are applied:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Selecting the correct senses",
"sec_num": "3.2"
},
{
"text": "Step 1. Obtaining the RST from candidate senses In this step we associate to each possible sense of each lemma a RST based on each semantic dimension. At this stage the aim of RST is to measure the relation between each Concept and each sense. To do this we use the Equation 2 where we have substituted the variable w (word) with the variable sw i , (sense) where sw i indicates the i-th sense of word w. As a result, we convert each RST in a vector. Next, we continue with the complete process adding the parent concepts.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Selecting the correct senses",
"sec_num": "3.2"
},
{
"text": "To measure the similarity between the RST of the sentences and senses, we have applied a fragment of the original method from (Guti\u00e9rrez et al., 2010a) introducing sense frequency ( ) as a new modification. Our goal is to obtain a new value to measure the Most Frequent Sense (MFS) in a given context. The AR value is accumulated when a matching exists between the vector elements of the sense and the vector elements of the sentence. The process is shown in the Equation 5.",
"cite_spans": [
{
"start": 126,
"end": 151,
"text": "(Guti\u00e9rrez et al., 2010a)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Step 2. Obtaining accumulated values of relevance for each resource and frequency sense",
"sec_num": null
},
{
"text": "(",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": ";",
"sec_num": null
},
{
"text": "Where AC is the AR value accumulated for the analyzed elements; ARV is the vector of relevant concepts of the sentence with the format: AC is calculated for each RST (or Relevant Vector) of each semantic dimension. In this approach we have obtained four AC values (for WN taxonomy, WND, WNA and SUMO).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": ";",
"sec_num": null
},
{
"text": "ARV[concept1 | AR",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": ";",
"sec_num": null
},
{
"text": "Notice that once we have obtained AC values for each sense in each dimension, if the senses calculated do not match with the grammatical category that Freeling (Atserias et al., 2006) suggests, we discriminate these senses adding a zero value to AC; in other case we add a one value. Adding these values we can maintain all the candidates in the solution despite the grammatical category is wrong.",
"cite_spans": [
{
"start": 160,
"end": 183,
"text": "(Atserias et al., 2006)",
"ref_id": "BIBREF13"
}
],
"ref_spans": [],
"eq_spans": [],
"section": ";",
"sec_num": null
},
{
"text": "Finally, the proposed sense will have the highest AC value among all senses in each lemma.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": ";",
"sec_num": null
},
{
"text": "Step",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": ";",
"sec_num": null
},
{
"text": "As we have explained above, each semantic dimension provides a possible sense. It is important to remark that the sense frequency is also included as a semantic dimension. So, in order to decide the right sense among the different semantic dimensions sense proposals we use a voting process. To apply this idea we define the next equation:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Voting process to obtain the final senses",
"sec_num": "3."
},
{
"text": "; (6)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Voting process to obtain the final senses",
"sec_num": "3."
},
{
"text": "Where VAC corresponds to a vector composed by AC values of each sense for one lemma; V [VAC] is a vector of the VAC; k corresponds to each resource; : corresponds to k-th VAC for resource k;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Voting process to obtain the final senses",
"sec_num": "3."
},
{
"text": ": determines the sense with maximum AC value of each VAC; i: is i-th sense;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Voting process to obtain the final senses",
"sec_num": "3."
},
{
"text": ": determines the sense that was selected more times by max k among all resources; and Ps: indicates proposed sense.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Voting process to obtain the final senses",
"sec_num": "3."
},
{
"text": "The obtain a tie or disjoin senses, the proposed sense will be the most frequent. We have chosen this option because of empirical studies have demonstrated that MFS works better than others (Molina et al., 2002) .",
"cite_spans": [
{
"start": 190,
"end": 211,
"text": "(Molina et al., 2002)",
"ref_id": "BIBREF0"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Voting process to obtain the final senses",
"sec_num": "3."
},
{
"text": "In this section our purpose is to confirm the hypothesis presented in Section 1. We have evaluated this method with two different test corpus, Senseval-2 on \"English All words\" task and Semeval-2 on \"English All words on Specific Domain\" task. Moreover, we have compared our results with the participating systems of the aforementioned competitions. The goal of these experiments is to demonstrate how the sense frequencies combined with RST can improve the original RST results.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluations and Analysis",
"sec_num": "4"
},
{
"text": "First, we analyzed how the addition of the sense frequencies to accumulated value (AC) of each sense improved the results of the previous work published on (Guti\u00e9rrez et al., 2010a) . To do this we used as test corpus the file d00.txt and we conducted some experiments: \uf0b7 Exp 1: Adding to AC value a 0% of Freq s . \uf0b7 Exp 2: Adding to AC value a 50% of Freq s . \uf0b7 Exp 3: Adding to AC value a 100% of Freq s . In the original method the authors calculated an accumulated value for each resource and summed up all the values to obtain the total accumulated value to combine all resources. In this new approach we also add the Freq s to the total accumulated value. Table 3 shows how each experiment obtains better results when Sense Frequencies (Freq s ) parameter is increased. Notice that we do not keep increasing this weight (i.e. 150%, 200%, etc) because the proposal would become converted only in selection process of MFS.",
"cite_spans": [
{
"start": 156,
"end": 181,
"text": "(Guti\u00e9rrez et al., 2010a)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [
{
"start": 662,
"end": 669,
"text": "Table 3",
"ref_id": "TABREF7"
}
],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "In order to determine whether the Freq s enhances the Most Frequent Senses (MFS) baseline, we conducted new experiments.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "Next, we show how we have used the original method adding to AC the 100% of Freq s but only using one dimension at the same time:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "\uf0b7 Exp4: Using MFS using Freq s \uf0b7 Exp5: Using WND resource \uf0b7 Exp6: Using SUMO resource \uf0b7 Exp7: Using WNA resource \uf0b7 Exp8: Using WN Taxonomy resource",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "After doing these experiments we were able to determine which dimension worked better. As we can see on Table 3 , these five experiments obtained promising results.",
"cite_spans": [],
"ref_spans": [
{
"start": 104,
"end": 111,
"text": "Table 3",
"ref_id": "TABREF7"
}
],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "Another experiment was to combine these five experiments in a voting process to obtain even better results. This idea has led us to make our main proposal.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "\uf0b7 Exp9: Applying a voting process among Exp4, Exp5, Exp6, Exp7 and Exp8 results Table 3 shows all the results obtained from d00.txt file of Senseval-2. The result of MFS is underlined and the approach that exceeded it is in bold. We can see that the voting process (Exp9) obtained the best results. Following, we present the results after analyzing the entire corpus of the Senseval-2 competition. For that, we applied two experiments to the entire corpus.",
"cite_spans": [],
"ref_spans": [
{
"start": 80,
"end": 87,
"text": "Table 3",
"ref_id": "TABREF7"
}
],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "\uf0b7 Exp10: Applying WSD with MFS of Freq s \uf0b7 Exp11: Applying a voting process using the five dimensions We show in Table 4 a comparison among the results of the best performances of our voting process, MFS using Freq s and MFS obtained by (Preiss, 2006) . The baseline used by Preiss was based on cntlist file from WN 1.7 version and our Exp10 was based on cntlist from WN 1.6. Notice, that are different although both are based on frequency information. As we can see, our approach improves the Exp10 results. These results were obtained by our system, but the baseline MFS results obtained by Preiss were better than ours. This means that we could enhance the MFS that we use. So, we need to integrate in our approach a better MFS resource to obtain better results. Table 4 shows that our proposal would have the best results of all unsupervised methods.",
"cite_spans": [
{
"start": 237,
"end": 251,
"text": "(Preiss, 2006)",
"ref_id": "BIBREF14"
}
],
"ref_spans": [
{
"start": 113,
"end": 120,
"text": "Table 4",
"ref_id": null
},
{
"start": 766,
"end": 773,
"text": "Table 4",
"ref_id": null
}
],
"eq_spans": [],
"section": "Evaluation with Senseval-2 corpus",
"sec_num": "4.1"
},
{
"text": "Our approach was also evaluated using corpus from Semeval-2 competition. The voting process obtained 52.7% and 51.5% of precision and recall respectively, improving the MFS baseline with 1% of accuracy. The original method from (Guti\u00e9rrez et al., 2010a) was improved on 19.3% of accuracy such as Table 5 shows. The underlined results pertain to original method from (Guti\u00e9rrez et al., 2010a) and the bold results pertain to our approach. As a result, we can see that we can improve the MFS proposal from Semeval-2 competition.",
"cite_spans": [
{
"start": 228,
"end": 253,
"text": "(Guti\u00e9rrez et al., 2010a)",
"ref_id": "BIBREF21"
},
{
"start": 366,
"end": 391,
"text": "(Guti\u00e9rrez et al., 2010a)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [
{
"start": 296,
"end": 303,
"text": "Table 5",
"ref_id": null
}
],
"eq_spans": [],
"section": "Evaluation with Semeval-2 corpus",
"sec_num": "4.2"
},
{
"text": "In this competition only were evaluated nouns and verbs. The behavior of our approach for each category was: nouns 54.4% of precision and 53.7% of recall, and verbs 49.4% of precision and 45.4% of recall. Each category is effective in comparison with the best results obtained on this competition.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation with Semeval-2 corpus",
"sec_num": "4.2"
},
{
"text": "In order to determine if the annotation of grammatical categories influences on the results, we discovered that the Freeling tool introduced a noise of 2.62% when detecting nouns and for verbs 8.20%. These analyses indicate that the results would be better using another more accurate tool.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Evaluation with Semeval-2 corpus",
"sec_num": "4.2"
},
{
"text": "In this section we present a comparison with some relevant WSD methods. We can mention those approaches using page-rank such as (Sinha and Mihalcea, 2007) , and (Agirre and Soroa, 2009 ). These proposals were tested using \"English All Words\" task corpus from Senseval-2. In both proposals, Page-Rank method has been used to determine the centrality of structural lexical network using the semantic relations of WordNet. Then, to disambiguate each word the most weighted sense was chosen. These approaches obtained 58.6% and 56.37% of recall respectively. Other significant work is the ACL 2004 paper by (Mc.Carthy et al., 2004) where the most frequent senses were obtained from a variety of resources (Reuters Corpus and SemCor Corpus), some of which provide domain information. This proposal obtained a 64% of precision in all-nouns task; this is just 3% higher than our results. However, we achieved better results than Mihalcea and Agirre exceeding them around 5%. This improvement could seem very poor but talking about WSD is a great step forward.",
"cite_spans": [
{
"start": 128,
"end": 154,
"text": "(Sinha and Mihalcea, 2007)",
"ref_id": "BIBREF17"
},
{
"start": 161,
"end": 184,
"text": "(Agirre and Soroa, 2009",
"ref_id": "BIBREF11"
},
{
"start": 603,
"end": 627,
"text": "(Mc.Carthy et al., 2004)",
"ref_id": "BIBREF9"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Comparison with newer works",
"sec_num": "4.3"
},
{
"text": "In this paper we have presented the hypothesis that for word-sense disambiguation it is important to obtain the Most Frequent Senses depending on the kind of analyzed context. In order to demonstrate this hypothesis, we have studied how several semantic dimensions combined with sense frequencies could improve the obtained results of many approaches that only conducted the WSD analysis with one dimension. We have proposed an adaptation of an unsupervised knowledge-based method that combines the original Relevant Semantic Trees method with senses frequency in a voting process. As a result, we have been able to determine which percentage of sense frequency is needed to help the Relevant Semantic Trees method. Therefore, we have demonstrated that the WSD results are better when more percentage of sense frequency is added.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusions and further works",
"sec_num": "5"
},
{
"text": "Moreover, we have conducted different experiments in order to know which semantic dimensions achieve better results. These experiments demonstrated that the Domain dimension (WND) and WordNet dimension (WN Taxonomy) worked better than MFS (Frequency dimension). Also, a voting process has been applied among all dimensions obtaining in Senseval-2 an of 60.9% and achieving the best results of all unsupervised systems. Furthermore, related to Semeval-2 our approach has improved the baseline MFS and the original RST method.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusions and further works",
"sec_num": "5"
},
{
"text": "As further work we propose to use other resources on the voting process in order to add more dimensions and also, use a better frequency resource. Apart from that, we also have considered to use another grammatical categorizer, in order to reduce the noise introduced by misclassifying words.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusions and further works",
"sec_num": "5"
},
{
"text": "Valenciana (grant no. PROMETEO/2009/119, ACOMP/2010/288 and ACOMP/2011/001).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusions and further works",
"sec_num": "5"
},
{
"text": "http://www.senseval.org",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "http://multiwordnet.fbk.eu/ 3 http://www.lsi.upc.es/~nlp/meaning/meaning.html",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "http://www.cse.unt.edu/~rada/downloads.html#semcor",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "This paper has been supported partially by Ministerio de Ciencia e Innovaci\u00f3n -Spanish Government (grant no. TIN2009-13391-C04-01), and Conselleria d'Educaci\u00f3n -Generalitat",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgements",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Word Sense Disambiguation using Statistical Models and WordNet",
"authors": [
{
"first": "Antonio",
"middle": [],
"last": "Molina",
"suffix": ""
},
{
"first": "Ferran",
"middle": [],
"last": "Pla",
"suffix": ""
},
{
"first": "Encarna",
"middle": [],
"last": "Segarra",
"suffix": ""
},
{
"first": "Lidia",
"middle": [],
"last": "Moreno",
"suffix": ""
}
],
"year": 2002,
"venue": "Proceedings of 3rd International Conference on Language Resources and Evaluation",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Antonio Molina, Ferran Pla, Encarna Segarra and Lidia Moreno. 2002. Word Sense Disambiguation using Statistical Models and WordNet. Proceedings of 3rd International Conference on Language Resources and Evaluation: Las Palmas de Gran Canaria.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Combining Knowledge-and Corpus-based Word-Sense-Disambiguation Methods",
"authors": [
{
"first": "Andr\u00e9s",
"middle": [],
"last": "Montoyo",
"suffix": ""
},
{
"first": "Armando",
"middle": [],
"last": "Su\u00e1rez",
"suffix": ""
},
{
"first": "German",
"middle": [],
"last": "Rigau",
"suffix": ""
},
{
"first": "Manuel",
"middle": [],
"last": "Palomar",
"suffix": ""
}
],
"year": 2005,
"venue": "Journal of Artificial Intelligence",
"volume": "23",
"issue": "",
"pages": "299--330",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Andr\u00e9s Montoyo, Armando Su\u00e1rez, German Rigau and Manuel Palomar. 2005. Combining Knowledge-and Corpus-based Word-Sense- Disambiguation Methods. Journal of Artificial Intelligence, 23:299-330.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "A SUMO-based Semantic Analysis for Knowledge Extraction",
"authors": [
{
"first": "Amal",
"middle": [],
"last": "Zouaq",
"suffix": ""
},
{
"first": "Michel",
"middle": [],
"last": "Gagnon",
"suffix": ""
},
{
"first": "Benoit",
"middle": [],
"last": "Ozell",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 4th Language & Technology Conference",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Amal Zouaq, Michel Gagnon and Benoit Ozell. 2009. A SUMO-based Semantic Analysis for Knowledge Extraction. Proceedings of the 4th Language & Technology Conference: Pozna\u0144, Poland.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Integrating Subject Field Codes into WordNet",
"authors": [
{
"first": "Bernardo",
"middle": [],
"last": "Magnini",
"suffix": ""
},
{
"first": "Gabriela",
"middle": [],
"last": "Cavaglia",
"suffix": ""
}
],
"year": 2000,
"venue": "Proceedings of Third International Conference on Language Resources and Evaluation (LREC-2000",
"volume": "",
"issue": "",
"pages": "1413--1418",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Bernardo Magnini and Gabriela Cavaglia. 2000. Integrating Subject Field Codes into WordNet. Proceedings of Third International Conference on Language Resources and Evaluation (LREC-2000): 1413--1418.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Using Domain Information for Word Sense Disambiguation",
"authors": [
{
"first": "Bernardo",
"middle": [],
"last": "Magnini",
"suffix": ""
},
{
"first": "Carlo",
"middle": [],
"last": "Strapparava",
"suffix": ""
},
{
"first": "Giovanni",
"middle": [],
"last": "Pezzulo",
"suffix": ""
},
{
"first": "Alfio",
"middle": [],
"last": "Gliozzo",
"suffix": ""
}
],
"year": 2008,
"venue": "Proceedings of the First International Conference on Emerging Trends in Engineering and Technology",
"volume": "",
"issue": "",
"pages": "1187--1191",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Bernardo Magnini, Carlo Strapparava, Giovanni Pezzulo and Alfio Gliozzo. 2008. Using Domain Information for Word Sense Disambiguation. Proceedings of the First International Conference on Emerging Trends in Engineering and Technology (icetet 2008): 1187-1191. Nagpur, India.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "SENSEVAL-3: Third International Workshop on the evaluation of System of the Semantic Analysis of Text",
"authors": [
{
"first": "Benjamin",
"middle": [],
"last": "Snyder",
"suffix": ""
},
{
"first": "Martha",
"middle": [],
"last": "Palmer",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Benjamin Snyder and Martha Palmer. 2004. The English All Word Task. SENSEVAL-3: Third International Workshop on the evaluation of System of the Semantic Analysis of Text: Barcelona, Spain.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "WordNet. An Electronic Lexical Database",
"authors": [
{
"first": "Christiane",
"middle": [],
"last": "Fellbaum",
"suffix": ""
}
],
"year": 1998,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Christiane Fellbaum. 1998. WordNet. An Electronic Lexical Database. The MIT Press: University of Cambridge.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "WordNet-Affect: an affective extension of WordNet",
"authors": [
{
"first": "Carlo",
"middle": [],
"last": "Strapparava",
"suffix": ""
},
{
"first": "Alessandro",
"middle": [],
"last": "Valitutti",
"suffix": ""
}
],
"year": 2004,
"venue": "Proceedings of the 4th International Conference on Language Resources and Evaluation (LREC",
"volume": "",
"issue": "",
"pages": "1083--1086",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Carlo Strapparava and Alessandro Valitutti. 2004. WordNet-Affect: an affective extension of WordNet. Proceedings of the 4th International Conference on Language Resources and Evaluation (LREC 2004): 1083-1086. Lisbon.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Integrating Conceptual Density with WordNet Domains and CALD Glosses for Noun Sense Disambiguation",
"authors": [
{
"first": "Davide",
"middle": [],
"last": "Buscaldi",
"suffix": ""
},
{
"first": "Paolo",
"middle": [],
"last": "Rosso",
"suffix": ""
},
{
"first": "Francesco",
"middle": [],
"last": "Masulli",
"suffix": ""
}
],
"year": 2005,
"venue": "Proceedings of Espa\u00f1a for Natural Language Processing",
"volume": "",
"issue": "",
"pages": "267--276",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Davide Buscaldi, Paolo Rosso and Francesco Masulli. 2005. Integrating Conceptual Density with WordNet Domains and CALD Glosses for Noun Sense Disambiguation. Proceedings of Espa\u00f1a for Natural Language Processing (ESTAL-2005): 267- 276. Alicante, Spain.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Finding Predominant Word Senses in Untagged Text",
"authors": [
{
"first": "Diana",
"middle": [],
"last": "Mc",
"suffix": ""
},
{
"first": "Rob",
"middle": [],
"last": "Carthy",
"suffix": ""
},
{
"first": "Julie",
"middle": [],
"last": "Koeling",
"suffix": ""
},
{
"first": "John",
"middle": [],
"last": "Weeds",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Carroll",
"suffix": ""
}
],
"year": 2004,
"venue": "Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL'04",
"volume": "",
"issue": "",
"pages": "279--286",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Diana Mc.Carthy, Rob Koeling, Julie Weeds and John Carroll. 2004. Finding Predominant Word Senses in Untagged Text. Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL'04): 279--286.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "SemEval-2010 task 17: All-words word sense disambiguation on a specific domain",
"authors": [
{
"first": "Eneko",
"middle": [],
"last": "Agirre",
"suffix": ""
},
{
"first": "Christiane",
"middle": [],
"last": "Oier L\u00f3pez De Lacalle",
"suffix": ""
},
{
"first": "Shu-Kai",
"middle": [],
"last": "Fellbaum",
"suffix": ""
},
{
"first": "Maurizio",
"middle": [],
"last": "Hsieh",
"suffix": ""
},
{
"first": "Monica",
"middle": [],
"last": "Tesconi",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Monachini",
"suffix": ""
}
],
"year": 2010,
"venue": "Proceedings of the 5th International Workshop on Semantic Evaluation",
"volume": "",
"issue": "",
"pages": "75--80",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Eneko Agirre, Oier L\u00f3pez De Lacalle, Christiane Fellbaum, Shu-Kai Hsieh, Maurizio Tesconi, Monica Monachini, Piek Vossen and Roxanne Segers. 2010. SemEval-2010 task 17: All-words word sense disambiguation on a specific domain. Proceedings of the 5th International Workshop on Semantic Evaluation: 75-80. Los Angeles, California.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "Personalizing PageRank for Word Sense Disambiguation",
"authors": [
{
"first": "Eneko",
"middle": [],
"last": "Agirre",
"suffix": ""
},
{
"first": "Aitor",
"middle": [],
"last": "Soroa",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 12th conference of the European chapter of the Association for Computational Linguistics",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Eneko Agirre and Aitor Soroa. 2009. Personalizing PageRank for Word Sense Disambiguation. Proceedings of the 12th conference of the European chapter of the Association for Computational Linguistics (EACL-2009): Athens, Greece.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Mapping WordNet to the SUMO Ontology",
"authors": [
{
"first": "Ian",
"middle": [],
"last": "Niles",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ian Niles. 2001. Mapping WordNet to the SUMO Ontology. Teknowledge Corporation.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "FreeLing 1.3: Syntactic and semantic services in an opensource NLP library",
"authors": [
{
"first": "Jordi",
"middle": [],
"last": "Atserias",
"suffix": ""
},
{
"first": "Bernardino",
"middle": [],
"last": "Casas",
"suffix": ""
},
{
"first": "Elisabet",
"middle": [],
"last": "Comelles",
"suffix": ""
},
{
"first": "Meritxell",
"middle": [],
"last": "Gonz\u00e1lez",
"suffix": ""
}
],
"year": 2006,
"venue": "Proceedings of LREC'06",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Jordi Atserias, Bernardino Casas, Elisabet Comelles, Meritxell Gonz\u00e1lez, Lluis Padr\u00f3 and Muntsa Padr\u00f3. 2006. FreeLing 1.3: Syntactic and semantic services in an opensource NLP library. Proceedings of LREC'06: Genoa, Italy.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "A detailed comparison of WSD systems:an analysis of the system answers for the Senseval-2 English all words task",
"authors": [
{
"first": "Judita",
"middle": [],
"last": "Preiss",
"suffix": ""
}
],
"year": 2006,
"venue": "Natural Language Engineering",
"volume": "12",
"issue": "3",
"pages": "209--228",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Judita Preiss. 2006. A detailed comparison of WSD systems:an analysis of the system answers for the Senseval-2 English all words task. Natural Language Engineering, 12(3):209-228.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Exploring the construction of semantic class classiers for WSD. Sociedad Espa\u00f1ola para el Procesamiento del Lenguaje Natural",
"authors": [
{
"first": "Luis",
"middle": [],
"last": "Villarejo",
"suffix": ""
},
{
"first": "Llu\u00eds",
"middle": [],
"last": "M\u00e1rquez",
"suffix": ""
},
{
"first": "German",
"middle": [],
"last": "Rigau",
"suffix": ""
}
],
"year": 2005,
"venue": "",
"volume": "35",
"issue": "",
"pages": "195--202",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Luis Villarejo, Llu\u00eds M\u00e1rquez and German Rigau. 2005. Exploring the construction of semantic class classiers for WSD. Sociedad Espa\u00f1ola para el Procesamiento del Lenguaje Natural, 35: 195-202.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Introduction to the Special Issue on Word Sense Disambiguation: The State of the Art",
"authors": [
{
"first": "Nancy",
"middle": [],
"last": "Ide",
"suffix": ""
},
{
"first": "Jean",
"middle": [],
"last": "V\u00e9ronis",
"suffix": ""
}
],
"year": 1998,
"venue": "Computational Linguistics",
"volume": "24",
"issue": "1",
"pages": "1--40",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Nancy Ide and Jean V\u00e9ronis. 1998. Introduction to the Special Issue on Word Sense Disambiguation: The State of the Art. Computational Linguistics, 24(1):1-40.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "Unsupervised Graph-based Word Sense Disambiguation Using Measures of Word Semantic Similarity",
"authors": [
{
"first": "Ravi",
"middle": [],
"last": "Sinha",
"suffix": ""
},
{
"first": "Rada",
"middle": [],
"last": "Mihalcea",
"suffix": ""
}
],
"year": 2007,
"venue": "Proceedings of the IEEE International Conference on Semantic Computing",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ravi Sinha and Rada Mihalcea. 2007. Unsupervised Graph-based Word Sense Disambiguation Using Measures of Word Semantic Similarity. Proceedings of the IEEE International Conference on Semantic Computing (ICSC 2007): Irvine, CA.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "English Lexical Sample SRL and All Words. Proceedings of the 4th International Workshop on Semantic Evaluations",
"authors": [
{
"first": "S",
"middle": [],
"last": "Sameer",
"suffix": ""
},
{
"first": "Edward",
"middle": [],
"last": "Pradhan",
"suffix": ""
},
{
"first": "Dmitriy",
"middle": [],
"last": "Loper",
"suffix": ""
},
{
"first": "Martha",
"middle": [
"Stone"
],
"last": "Dligach",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Palmer",
"suffix": ""
}
],
"year": 2007,
"venue": "",
"volume": "",
"issue": "",
"pages": "87--92",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sameer S. Pradhan, Edward Loper, Dmitriy Dligach and Martha Stone Palmer. 2007. SemEval-2007 Task-17: English Lexical Sample SRL and All Words. Proceedings of the 4th International Workshop on Semantic Evaluations (SemEval- 2007): 87-92. Prague.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "English All word. SENSEVAL-2: Second International Workshop on Evaluating Word Sense Disambiguation Systems",
"authors": [
{
"first": "Scott",
"middle": [],
"last": "Cotton",
"suffix": ""
},
{
"first": "Phil",
"middle": [],
"last": "Edmonds",
"suffix": ""
},
{
"first": "Adam",
"middle": [],
"last": "Kilgarriff",
"suffix": ""
},
{
"first": "Martha",
"middle": [],
"last": "Palmer",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Scott Cotton, Phil Edmonds, Adam Kilgarriff and Martha Palmer. 2001. English All word. SENSEVAL-2: Second International Workshop on Evaluating Word Sense Disambiguation Systems: Toulouse, France.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Using Relevant Domains Resource for Word Sense Disambiguation. IC-AI'04",
"authors": [
{
"first": "Sonia",
"middle": [],
"last": "V\u00e1zquez",
"suffix": ""
},
{
"first": "Andr\u00e9s",
"middle": [],
"last": "Montoyo",
"suffix": ""
},
{
"first": "German",
"middle": [],
"last": "Rigau",
"suffix": ""
}
],
"year": 2004,
"venue": "Proceedings of the International Conference on Artificial Intelligence",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sonia V\u00e1zquez, Andr\u00e9s Montoyo and German Rigau. 2004. Using Relevant Domains Resource for Word Sense Disambiguation. IC-AI'04. Proceedings of the International Conference on Artificial Intelligence: Ed: CSREA Press. Las Vegas, E.E.U.U.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "UMCC-DLSI: Integrative resource for disambiguation task",
"authors": [
{
"first": "Yoan",
"middle": [],
"last": "Guti\u00e9rrez",
"suffix": ""
},
{
"first": "Antonio",
"middle": [],
"last": "Fern\u00e1ndez",
"suffix": ""
}
],
"year": 2010,
"venue": "Proceedings of the 5th International Workshop on Semantic Evaluation",
"volume": "",
"issue": "",
"pages": "427--432",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yoan Guti\u00e9rrez, Antonio Fern\u00e1ndez, Andr\u00e9s Montoyo and Sonia V\u00e1zquez. 2010a. UMCC-DLSI: Integrative resource for disambiguation task. Proceedings of the 5th International Workshop on Semantic Evaluation: 427-432. Uppsala, Sweden.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Integration of semantic resources based on WordNet. XXVI Congreso de la Sociedad Espa\u00f1ola para el Procesamiento del Lenguaje Natural",
"authors": [
{
"first": "Yoan",
"middle": [],
"last": "Guti\u00e9rrez",
"suffix": ""
},
{
"first": "Antonio",
"middle": [],
"last": "Fern\u00e1ndez",
"suffix": ""
}
],
"year": 2010,
"venue": "",
"volume": "45",
"issue": "",
"pages": "161--168",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yoan Guti\u00e9rrez, Antonio Fern\u00e1ndez, Andr\u00e9s Montoyo and Sonia V\u00e1zquez. 2010b. Integration of semantic resources based on WordNet. XXVI Congreso de la Sociedad Espa\u00f1ola para el Procesamiento del Lenguaje Natural, 45: 161-168. Universidad Polit\u00e9cnica de Valencia, Valencia.",
"links": null
}
},
"ref_entries": {
"TABREF2": {
"text": "]",
"content": "<table><tr><td/><td>Vector</td></tr><tr><td>AR Domain</td><td>AR Domain</td></tr><tr><td>0.90 Pedagogy</td><td>0.36 Commerce</td></tr><tr><td colspan=\"2\">0.90 Administration 0.36 Quality</td></tr><tr><td>0.36 Buildings</td><td>0.36 Psychoanalysis</td></tr><tr><td>0.36 Politics</td><td>0.36 Economy</td></tr><tr><td>0.36 Environment</td><td/></tr><tr><td colspan=\"2\">Table 1. Initial Vector of Domain</td></tr></table>",
"type_str": "table",
"num": null,
"html": null
},
"TABREF3": {
"text": "This bottom-up process is applied for each Concept of the Initial Vector to add each Relevant Parent to the vector. After reproducing the process to each Concept of the Initial Vector, the Relevant Semantic Tree is built. As a result, theTable 2is obtained. This vector represents the Domain tree associated to the sentence such as",
"content": "<table><tr><td>){</td></tr><tr><td>if ( PC had not been added to vector)</td></tr><tr><td>PC is added to the vector with AR(PC, s)</td></tr><tr><td>value;</td></tr><tr><td>else PC value = PC value + AR(PC, s) value; }</td></tr></table>",
"type_str": "table",
"num": null,
"html": null
},
"TABREF4": {
"text": "Final Domain Vector based on WND Figure 1. Relevant Semantic Tree from WNDOnce the Relevant Semantic Tree is obtained, in case of the Domain dimension the Factotum category is eliminated from the tree. Due to the fact that Factotum is a generic Domain associated to words that appear in general contexts it does not provide useful information",
"content": "<table><tr><td colspan=\"2\">Vector</td></tr><tr><td>AR Domain</td><td>AR Domain</td></tr><tr><td colspan=\"2\">1.63 Social_Science 0.36 Buildings</td></tr><tr><td colspan=\"2\">0.90 Administration 0.36 Commerce</td></tr><tr><td>0.90 Pedagogy</td><td>0.36 Environment</td></tr><tr><td>0.80 RootDomain</td><td>0.11 Factotum</td></tr><tr><td colspan=\"2\">0.36 Psychoanalysis 0.11 Psychology</td></tr><tr><td>0.36 Economy</td><td>0.11 Architecture</td></tr><tr><td>0.36 Quality</td><td>0.11 Pure_Science</td></tr><tr><td>0.36 Politics</td><td/></tr></table>",
"type_str": "table",
"num": null,
"html": null
},
"TABREF5": {
"text": "Vs[concepts]; Vs k is the k-th concept of the vector Vs; ARV [Vs k ] represents the value of AR assigned to the concept Vs k for the value ARV; Freq s represents the normalized value of frequency sense obtained from cntlist file from WN 1.6; and is the term that normalizes the result.",
"content": "<table/>",
"type_str": "table",
"num": null,
"html": null
},
"TABREF7": {
"text": "Results over d00.txt from Senseval-2",
"content": "<table/>",
"type_str": "table",
"num": null,
"html": null
}
}
}
} |