File size: 104,335 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 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 | {
"paper_id": "R13-1002",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T14:55:21.982056Z"
},
"title": "Optimising Tree Edit Distance with Subtrees for Textual Entailment",
"authors": [
{
"first": "Maytham",
"middle": [],
"last": "Alabbas",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "Basrah University",
"location": {
"settlement": "Basrah",
"country": "Iraq"
}
},
"email": "maytham.alabbas@gmail.com"
},
{
"first": "Allan",
"middle": [],
"last": "Ramsay",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "University of Manchester",
"location": {
"postCode": "M13 9PL",
"settlement": "Manchester",
"country": "UK"
}
},
"email": "allan.ramsa@manchester.ac.uk"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "This paper introduces a method for improving tree edit distance (TED) for textual entailment. We explore two ways of improving TED: we extend the standard TED to use edit operations that apply to subtrees as well as to single nodes; and we use the 'artificial bee colony' algorithm (ABC) to estimate the cost of edit operations for single nodes and subtrees and to determine thresholds. The preliminary results of the current work for checking entailment between two texts are encouraging compared with the common bag-ofwords, string edit distance and standard TED algorithms.",
"pdf_parse": {
"paper_id": "R13-1002",
"_pdf_hash": "",
"abstract": [
{
"text": "This paper introduces a method for improving tree edit distance (TED) for textual entailment. We explore two ways of improving TED: we extend the standard TED to use edit operations that apply to subtrees as well as to single nodes; and we use the 'artificial bee colony' algorithm (ABC) to estimate the cost of edit operations for single nodes and subtrees and to determine thresholds. The preliminary results of the current work for checking entailment between two texts are encouraging compared with the common bag-ofwords, string edit distance and standard TED algorithms.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "One key task for natural language systems is to determine whether one natural language sentence entails another. Entailment can be defined as a relationship between two sentences where the truth of one sentence, the entailing expression, forces the truth of another sentence, what is entailed. Many natural language processing (NLP) tasks such as information extraction and question answering have to cope with this notion.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "An alternative formulation for the entailment between two texts is given by the recognising textual entailment (RTE) paradigm, which contrasts with the standard definition of entailment above. Dagan et al. (2005) describe RTE as a task of determining, for two sentences text T and hypothesis H, whether \". . . typically, a human reading T would infer that H is most likely true.\" According to these authors, entailment holds if the truth of H, as interpreted by a typical language user, can be inferred from the meaning of T. This notion of entailment is less rigorous, and less clearly defined, than the standard notion, but it can be useful for a number of tasks, and has been investigated very extensively in recent times.",
"cite_spans": [
{
"start": 193,
"end": 212,
"text": "Dagan et al. (2005)",
"ref_id": "BIBREF8"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Tree edit distance (TED), which models T-H pairs by explicitly transforming T into H via a minimal cost sequence of editing operations, has been widely used for this task. Using TED poses two challenges: the standard three operations (i.e. deletion, insertion and exchange) apply only to single nodes, rather than to subtrees; and estimating a combination of costs for these operations with threshold(s) is hard when dealing with complex problems. This is because alterations in these costs or choosing a different combination of them can lead to drastic changes in TED performance (Mehdad and Magnini, 2009) .",
"cite_spans": [
{
"start": 582,
"end": 608,
"text": "(Mehdad and Magnini, 2009)",
"ref_id": "BIBREF24"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "In order to overcome these challenges, we have extended the standard TED to deal with subtree operations as well as operations on single nodes. This allows the algorithm to treat semantically coherent parts of the tree as single items, thus allowing for instance entire modifiers (such as prepositional phrase (PPs)) to be inserted or deleted as single units. We have also applied the artificial bee colony (ABC) algorithm (Akay and Karaboga, 2012) to estimate costs both of edit operations (single node and subtree) and of threshold(s).",
"cite_spans": [
{
"start": 423,
"end": 448,
"text": "(Akay and Karaboga, 2012)",
"ref_id": "BIBREF0"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The work was carried out as part of an attempt to build a textual entailment (TE) system for modern standard Arabic (MSA) (Alabbas, 2011) . MSA poses a number of problems that, while familiar from other languages, make tasks such as TE particularly difficult for this language-the lack of diacritics in written MSA combines with the complex derivational and inflectional morphology of the language to produce worse levels of lexical ambiguity than occur in many other languages; the combination of free word-order, pro-drop, verbless sentences and complex nominals produces higher levels of syntactic ambiguity than occur in many other languages; and the combination of these combinations makes things even worse. We have tested our algorithms on a corpus of MSA T-H pairs. This corpus contains 600 pairs, binary annotated as 'yes' and 'no' (a 50%-50% split). The average length of sentence in this dataset is 25 words per sentence, with some sentences containing 40+ words (see (Alabbas, 2013) for further details of this dataset and description of the methodology used for collecting it). In order to maintain comparability with work on TE for English, in Section 4 we have replicated a number of standard techniques (bag-of-words, Levenshtein distance on strings, standard TED). These experiments show that the extended version of TED, ETED, improves the performance of our technique for Arabic by around 3% in f-score and around 2% in accuracy compared with a number of wellknown techniques. The relative performance of the standard techniques on our Arabic testset replicates the results reported for these techniques for English testsets. We have also applied our ETED to the English RTE2 testset, where it again outperforms the standard version of TED.",
"cite_spans": [
{
"start": 122,
"end": 137,
"text": "(Alabbas, 2011)",
"ref_id": "BIBREF3"
},
{
"start": 979,
"end": 994,
"text": "(Alabbas, 2013)",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The idea here is to convert both T and H from natural language expressions into parse trees through parsing and then to explicitly transform T's parse tree into H's parse tree, using a sequence of edit operations (Kouylekov and Magnini, 2005; Bar-Haim et al., 2007; Harmeling, 2009; Mehdad and Magnini, 2009; Wang and Manning, 2010; Heilman and Smith, 2010; Stern et al., 2012) . If a lowcost transformation sequence can be found then it may be that T entails H. Dependency parsers (K\u00fcbler et al., 2009 ) are popular for this task, as in other NLP areas in recent years, since they allow us to be sensitive to the fact that the links in a dependency tree carry linguistic information about relations between complex units.",
"cite_spans": [
{
"start": 213,
"end": 242,
"text": "(Kouylekov and Magnini, 2005;",
"ref_id": "BIBREF19"
},
{
"start": 243,
"end": 265,
"text": "Bar-Haim et al., 2007;",
"ref_id": "BIBREF5"
},
{
"start": 266,
"end": 282,
"text": "Harmeling, 2009;",
"ref_id": "BIBREF12"
},
{
"start": 283,
"end": 308,
"text": "Mehdad and Magnini, 2009;",
"ref_id": "BIBREF24"
},
{
"start": 309,
"end": 332,
"text": "Wang and Manning, 2010;",
"ref_id": "BIBREF33"
},
{
"start": 333,
"end": 357,
"text": "Heilman and Smith, 2010;",
"ref_id": "BIBREF13"
},
{
"start": 358,
"end": 377,
"text": "Stern et al., 2012)",
"ref_id": "BIBREF32"
},
{
"start": 482,
"end": 502,
"text": "(K\u00fcbler et al., 2009",
"ref_id": "BIBREF20"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "TED for RTE",
"sec_num": "2"
},
{
"text": "Different sets of operations on trees, using various types of transformations in order to derive H from T, have been suggested. Herrera et al. (2005) , for instance, used the notion of tree inclusion (Kilpel\u00e4inen, 1992) , which obtained one tree from another by deleting nodes. Herrera et al. (2006) and Marsi et al. (2006) used a tree alignment algorithm (Meyers et al., 1996) , which produces a multiple sequence alignment on a set of sequences over a fixed tree. TED (Zhang and Shasha, 1989; Klein et al., 2000; Pawlik and Augsten, 2011) is another example of a transformation-based model in that it computes the minimum cost sequence of transformations (e.g. insertion, deletion and exchange of nodes) that turns one tree into the other. To obtain more accurate predictions, it is important to define an appropriate inventory of edit operations and assign appropriate costs to the edit operations during a training stage (Kouylekov and Magnini, 2005; Harmeling, 2009) . For instance, exchanging a noun with its synonyms or hypernyms should cost less than exchanging it with an unrelated word. Heilman and Smith (2010) extended the above mentioned operations (e.g. move-sibling, relabel-edge, move-subtree, etc.), since the available edit operations are limited in capturing certain interesting and prevalent semantic phenomena. Similarly, a heuristic set of 28 edit operations, which include numbers of nodeexchanges and restructuring of the entire parse tree, is suggested (Harmeling, 2009) .",
"cite_spans": [
{
"start": 128,
"end": 149,
"text": "Herrera et al. (2005)",
"ref_id": "BIBREF14"
},
{
"start": 200,
"end": 219,
"text": "(Kilpel\u00e4inen, 1992)",
"ref_id": "BIBREF16"
},
{
"start": 278,
"end": 299,
"text": "Herrera et al. (2006)",
"ref_id": "BIBREF15"
},
{
"start": 304,
"end": 323,
"text": "Marsi et al. (2006)",
"ref_id": "BIBREF22"
},
{
"start": 356,
"end": 377,
"text": "(Meyers et al., 1996)",
"ref_id": "BIBREF25"
},
{
"start": 470,
"end": 494,
"text": "(Zhang and Shasha, 1989;",
"ref_id": "BIBREF34"
},
{
"start": 495,
"end": 514,
"text": "Klein et al., 2000;",
"ref_id": "BIBREF17"
},
{
"start": 515,
"end": 540,
"text": "Pawlik and Augsten, 2011)",
"ref_id": "BIBREF27"
},
{
"start": 925,
"end": 954,
"text": "(Kouylekov and Magnini, 2005;",
"ref_id": "BIBREF19"
},
{
"start": 955,
"end": 971,
"text": "Harmeling, 2009)",
"ref_id": "BIBREF12"
},
{
"start": 1097,
"end": 1121,
"text": "Heilman and Smith (2010)",
"ref_id": "BIBREF13"
},
{
"start": 1478,
"end": 1495,
"text": "(Harmeling, 2009)",
"ref_id": "BIBREF12"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "TED for RTE",
"sec_num": "2"
},
{
"text": "TED-based inference requires the specification of a cost for each edit operation and a threshold for the total cost of the edit sequence. Selecting a best set of costs and a suitable threshold is challenging. Some researchers have defined costs manually (Kouylekov and Magnini, 2005) , but they are usually learned automatically (Harmeling, 2009; Wang and Manning, 2010; Heilman and Smith, 2010; Stern and Dagan, 2011) , e.g. Mehdad and Magnini (2009) have used particle swarm optimization (PSO), which is a stochastic technique that mimics the social behaviour of bird flocking and fish schooling (Russell and Cohn, 2012) , for estimating and optimising the cost of each edit operation for TED.",
"cite_spans": [
{
"start": 254,
"end": 283,
"text": "(Kouylekov and Magnini, 2005)",
"ref_id": "BIBREF19"
},
{
"start": 329,
"end": 346,
"text": "(Harmeling, 2009;",
"ref_id": "BIBREF12"
},
{
"start": 347,
"end": 370,
"text": "Wang and Manning, 2010;",
"ref_id": "BIBREF33"
},
{
"start": 371,
"end": 395,
"text": "Heilman and Smith, 2010;",
"ref_id": "BIBREF13"
},
{
"start": 396,
"end": 418,
"text": "Stern and Dagan, 2011)",
"ref_id": "BIBREF31"
},
{
"start": 426,
"end": 451,
"text": "Mehdad and Magnini (2009)",
"ref_id": "BIBREF24"
},
{
"start": 598,
"end": 622,
"text": "(Russell and Cohn, 2012)",
"ref_id": "BIBREF30"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "TED for RTE",
"sec_num": "2"
},
{
"text": "In this paper we will use Zhang and Shasha (1989) 's TED algorithm (henceforth, ZS-TED), which is an efficient technique based on dynamic programming to calculate the approximate tree matching for two rooted ordered trees, as a starting point. Ordered trees are trees in which the left-toright order among siblings is significant. Approximate tree matching allows us to match a complete tree with just some parts of another tree. There are three operations, namely deleting, inserting and exchanging a node, which can transform one ordered tree to another. A nonnegative real cost is associated with each edit operation. These costs are changed to match the requirements of specific applications. Deleting a node x means attaching its children to the parent of x. Insertion is the inverse of deletion, with an inserted node becoming a parent of a consecutive subsequence in the left-to-right order of its parent. Exchanging a node alters its label. Detailed presentation of ZS-TED can be found in (Bille, 2005) : the main change that we make to the basic algorithm is to include extra tables for recording which operations were performed rather than simply recording their cost.",
"cite_spans": [
{
"start": 26,
"end": 49,
"text": "Zhang and Shasha (1989)",
"ref_id": "BIBREF34"
},
{
"start": 997,
"end": 1010,
"text": "(Bille, 2005)",
"ref_id": "BIBREF6"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Standard TED",
"sec_num": "2.1"
},
{
"text": "The main weakness of ZS-TED is that it is not able to perform transformations on subtrees (i.e. delete subtree, insert subtree and exchange subtree). In order to make ZS-TED deal with subtree operations, we need to follow two stages:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Extended TED",
"sec_num": "2.2"
},
{
"text": "1. Run ZS-TED (without entire subtree operations) and compute the standard alignment from the results;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Extended TED",
"sec_num": "2.2"
},
{
"text": "2. Go over the alignment and group subtrees operations (e.g. every consecutive k deletions that correspond to an entire subtree reduces the edit distance score by \u03b1 \u00d7 k + \u03b2 for any desired \u03b1 and \u03b2 in interval [0,1]).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Extended TED",
"sec_num": "2.2"
},
{
"text": "We have applied this technique on Zhang and Shasha (1989)'s O(n 4 ) algorithm but it will also work for Klein (1998) 's O(n 3 log n ) algorithm, Demaine et al. (2009)'s O(n 3 ) algorithm or Pawlik and Augsten (2011)'s O(n 3 ) algorithm. The additional time cost of O(n 2 ) can be ignored since it is less than the time cost for any available TED algorithm.",
"cite_spans": [
{
"start": 104,
"end": 116,
"text": "Klein (1998)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Extended TED",
"sec_num": "2.2"
},
{
"text": "In order to find the sequence of edit operations that transforms one tree into another, such as the pair shown in Figure 1 , the computation proceeds as follows: create a new matrix called \u03b4 2 , which has the same dimensions as the matrix \u03b4 which is used to store the forest costs during ZS-TED to store the sequence of edit operations as a list. In particular, when the values of \u03b4 are computed, the values of \u03b4 2 are computed, by using the edit operation labels: \"i\" for an insertion, \"d\" for deletion, \"x\" for exchange and \"m\" for no operation (matching). So, the final edit sequence to transform T 1 into T 2 in Figure 1 is dddmmiiimm.",
"cite_spans": [],
"ref_spans": [
{
"start": 114,
"end": 122,
"text": "Figure 1",
"ref_id": "FIGREF0"
},
{
"start": 616,
"end": 624,
"text": "Figure 1",
"ref_id": "FIGREF0"
}
],
"eq_spans": [],
"section": "Find a sequence of single operations",
"sec_num": "2.2.1"
},
{
"text": "The final mapping between T 1 and T 2 is shown in Figure 1 ",
"cite_spans": [],
"ref_spans": [
{
"start": 50,
"end": 58,
"text": "Figure 1",
"ref_id": "FIGREF0"
}
],
"eq_spans": [],
"section": "Find a sequence of single operations",
"sec_num": "2.2.1"
},
{
"text": "Extending TED to cover subtree operations will give us more flexibility when comparing trees (especially linguistic trees). Thus, we have extended the TED algorithm to allow the standard edit operations (insert, delete and exchange) to apply both single nodes and subtrees. Let E p=1..L \u2208 {\"d\", \"i\", \"x\", \"m\"} be an edit operation sequence that transforms T 1 into T 2 by applying the technique in Section 2.2.1. Suppose that S 1 and S 2 are the optimal alignment for T 1 and T 2 respectively, when the length of",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Find a sequence of subtree operations",
"sec_num": "2.2.2"
},
{
"text": "S 1 = S 2 = L.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Find a sequence of subtree operations",
"sec_num": "2.2.2"
},
{
"text": "To find the optimal single and subtree edit operations sequence that transform T 1 into T 2 , each largest sequence of same operation is checked to see whether it contains subtree(s) or not. Checking whether such a sequence corresponds to a subtree depends on the type of edit operation, according to the following rules: (i) if the operation is \"d,\" the sequence is checked on the first tree; (ii) if the operation is \"i,\" the sequence is checked on the second tree; and (iii) otherwise, the sequence is checked on both trees. After that, if the sequence of operations corresponds to a subtree, then all the symbols of the sequence are replaced by \"+\" except the last one (which represents the root of the subtree). Otherwise, checking starts from a new sequence as explained below. For instance, let us consider E h , ..., E t , where 1 \u2264 h < L, 1 < t \u2264 L, h < t, is a sequence of the same edit operation, i.e. E k=h..t \u2208 {\"d\", \"i\", \"x\", \"m\"}. Let us consider h0 = h, we firstly check nodes S 1 h , ..., S 1 t and S 2 h , ..., S 2 t to see whether they or not are subtrees. If E k is \"d,\" the nodes S 1 h , ..., S 1 t are checked, whereas the nodes S 2 h , ..., S 2 t are checked when E k is \"i.\" Otherwise, the nodes S 1 h , ..., S 1 t and S 2 h , ..., S 2 t are checked. All edit operations E h , ..., E t\u22121 are replaced by \"+\" when this sequence is corresponding to a subtree. Then, we start checking from the beginning of another sequence from the left of the subtree E h , ..., E t , i.e. t = h \u2212 1. Otherwise, the checking is applied with the sequence start from the next position, i.e. h = h + 1. The checking is continued until h = t. After that, when the (t \u2212 h) sequences that start with different positions and end with t position do not contain a subtree, the checking starts from the beginning with the new sequence, i.e. h = h0 and t = t \u2212 1. The process is repeated until h = t.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Find a sequence of subtree operations",
"sec_num": "2.2.2"
},
{
"text": "So, the final edit sequence to transform T 1 into T 2 in Figure 1 is ++d+m++imm.",
"cite_spans": [],
"ref_spans": [
{
"start": 57,
"end": 65,
"text": "Figure 1",
"ref_id": "FIGREF0"
}
],
"eq_spans": [],
"section": "Find a sequence of subtree operations",
"sec_num": "2.2.2"
},
{
"text": "The final mapping between T 1 and T 2 according to the extended TED is shown in Figure 2 . ",
"cite_spans": [],
"ref_spans": [
{
"start": 80,
"end": 88,
"text": "Figure 2",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "Find a sequence of subtree operations",
"sec_num": "2.2.2"
},
{
"text": "We used two optimisation algorithms, genetic algorithm (GA) and artificial bee colony (ABC), to estimate the cost of each edit operation (i.e. for single nodes and for subtrees) and threshold(s) based on application and type of system output.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Optimisation algorithms",
"sec_num": "3"
},
{
"text": "The GA starts with an initial population of solutions (known as chromosomes). In each generation, solutions from the current population are taken and used to form a new population by modifying the selected solutions' genome (recombined and possibly randomly mutated). This is motivated by a hope that the new population will be better than the old one. Solutions which are selected to form new solutions (offspring) are se-lected according to their fitness-the more suitable they are the more chances they have to reproduce. The algorithm terminates when either a maximum number of generations has been produced, or a satisfactory fitness level has been reached for the population. The main steps of the algorithm are shown in Algorithm 1.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "GA",
"sec_num": "3.1"
},
{
"text": "The basic algorithm for GA.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Algorithm 1",
"sec_num": null
},
{
"text": "1: Initialise population; 2: repeat 3: Evaluation; 4:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Algorithm 1",
"sec_num": null
},
{
"text": "Reproduction; 5:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Algorithm 1",
"sec_num": null
},
{
"text": "Crossover; 6: Mutation; 7: until (termination conditions are met);",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Algorithm 1",
"sec_num": null
},
{
"text": "In the ABC algorithm, the colony of artificial bees consists of three groups. First, employed bees going to the food source (a possible solution to the problem to be optimised) that they have visited previously. Second, onlookers waiting to choose a food source. Third, scouts carrying out random search. The first half of the colony consists of the employed artificial bees and the second half includes the onlookers and scouts. The number of employed bees is equal to the number of food sources. The employed bee of an abandoned food source becomes a scout. The main steps of the algorithm are shown in Algorithm 2.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "ABC algorithm",
"sec_num": "3.2"
},
{
"text": "ABC follows three steps during each cycle: (i) moving both the employed and onlooker bees onto the food sources; (ii) calculating their nectar amounts (fitness value); and (iii) determining the scout bees and then moving them randomly onto the possible food sources.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "ABC algorithm",
"sec_num": "3.2"
},
{
"text": "The ABC algorithm has been widely used in many optimisation applications, since it is easy to implement and has fewer control parameters.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "ABC algorithm",
"sec_num": "3.2"
},
{
"text": "To check the effectiveness of the extended TED with subtree operations, ETED, we used it to check the entailment between T-H Arabic pairs of text snippets and compared its results with a simple bag-of-words, Levenshtein distance and ZS-TED on the same set of pairs.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental results",
"sec_num": "4"
},
{
"text": "We have investigated different approaches that can be divided into two groups as follow.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "Algorithm 2 Pseudo-code of the ABC algorithm (Akay and Karaboga, 2012) . Apply the greedy selection process for the employed bees (if the new solution vij has an equal or better nectar (fitness) than the old source, it is replaced with the old one in the memory. Otherwise, the old one is retained in the memory); 7:",
"cite_spans": [
{
"start": 45,
"end": 70,
"text": "(Akay and Karaboga, 2012)",
"ref_id": "BIBREF0"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "Calculate the probability values pi = fiti/ SN i=1 fiti for the solutions xi; 8:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "Produce the new solutions vij for the onlookers from the solutions xi selected depending on pi and evaluate them; 9:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "Apply the greedy selection process for the onlookers; 10:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "Determine the abandoned solution for the scout, if exists, and replace it with a new randomly produced solution xi by",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "x j i = x j min +rand[0,1](x j max -x j min ); 11:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "Memorise the best solution achieved so far; 12: cycle = cycle+1; 13: until (cycle = Maximum Cycle Number);",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Systems",
"sec_num": "4.1"
},
{
"text": "We tested the following approaches:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Surface string similarity approaches",
"sec_num": null
},
{
"text": "BoW: this approach uses the bag-of-words, which measures the similarity between T and H as a number of common words between them (either in surface forms or lemma forms), divided by the length of H, when the highest similarity is better.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Surface string similarity approaches",
"sec_num": null
},
{
"text": "LD 1 : this approach uses the Levenshtein distance with 0.5, 1, 1.5 for cost of deleting, inserting and exchanging a word respectively. LD 2 : the same as for LD 1 except that the cost of exchanging non-identical words is the Levenshtein distance between the two words (with lower costs for vowels) divided by the length of the longer of the two words (derived and inflected forms of Arabic words tend to share the same consonants, at least in the root, so this provides a very approximate solution to the task of determining whether two forms correspond to the same lexical item).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Surface string similarity approaches",
"sec_num": null
},
{
"text": "These approaches follow three steps:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "1. each sentence is preprocessed by a tagger and a parser in order to convert them to dependency trees, using a combination of taggers (i.e. AMIRA (Diab, 2009) , MADA (Habash et al., 2009) and maximum-likelihood (MXL) tagger (Ramsay and Sabtan, 2009) ) and parsers (i.e. MALTParser (Nivre et al., 2007) and MST-Parser (McDonald et al., 2006) ), which give around 85% for labelled accuracy (Alabbas and Ramsay, 2012; Alabbas and Ramsay, 2011) , which is the best result we have seen for the Penn Arabic treebank (PATB). We use these combinations in series of experiments which involve;",
"cite_spans": [
{
"start": 147,
"end": 159,
"text": "(Diab, 2009)",
"ref_id": "BIBREF10"
},
{
"start": 167,
"end": 188,
"text": "(Habash et al., 2009)",
"ref_id": "BIBREF11"
},
{
"start": 225,
"end": 250,
"text": "(Ramsay and Sabtan, 2009)",
"ref_id": "BIBREF29"
},
{
"start": 282,
"end": 302,
"text": "(Nivre et al., 2007)",
"ref_id": "BIBREF26"
},
{
"start": 307,
"end": 341,
"text": "MST-Parser (McDonald et al., 2006)",
"ref_id": null
},
{
"start": 389,
"end": 415,
"text": "(Alabbas and Ramsay, 2012;",
"ref_id": "BIBREF2"
},
{
"start": 416,
"end": 441,
"text": "Alabbas and Ramsay, 2011)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "2. pairs of dependency trees are matched using the ZS-TED/ETED to obtain a score for the pair;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "3. either one threshold (for simple entails/fails-toentail tests or two (for entails/unknown/fails-toentail tests) are used to determine whether this score should lead to a particular judgement.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "We tested the following approaches:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "ZS-TED 1 : this system uses ZS-TED with a manually determined set of fixed costs. The cost of deleting a node, inserting a node or exchanging a node are 0, 10 and 10 respectively. ZS-TED 2 : this system uses ZS-TED with a manually determined intuition-based set of costs that depend on a set of stopwords and on sets of synonyms and hypernyms, obtained from Arabic WordNet (AWN) (Black et al., 2006) , as explained in Figure 3 (column A). These costs are an updated version of the costs used by Punyakanok et al. (2004) .",
"cite_spans": [
{
"start": 379,
"end": 399,
"text": "(Black et al., 2006)",
"ref_id": "BIBREF7"
},
{
"start": 495,
"end": 519,
"text": "Punyakanok et al. (2004)",
"ref_id": "BIBREF28"
}
],
"ref_spans": [
{
"start": 418,
"end": 426,
"text": "Figure 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "ZS-TED+GA: this system uses a GA to estimate the costs of edit single operations and threshold(s) for ZS-TED. The chromosome for binary decision output is {cost of deleting a node, cost of inserting a node, cost of exchanging a node, threshold}, and the fitness is a*f-score+b*accuracy, where a and b are real numbers in the interval [0,1]. Providing different values for a and b makes it possible to optimise the system for different applications-in the experiments below a is 0.6 and b is 0.4, which effectively puts more emphasis on precision than on recall, but for other tasks different values could be used. For three-way decisions, the chromosome is the same as for binary decisions except that we add a second threshold, and the fitness is simply the f-score. We used the steady state GA with the following settings: 40 chromosomes as population size, uniform crossover (UX), Gaussian mutation and maximum number of generations is 100.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "ZS-TED+ABC: the same as ZS-TED+GA except using ABC instead of GA as the optimisation algorithm. We used the ABC algorithm with the following settings: 40 as the colony size and the maximum number of cycles for foraging is 100.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "ETED 1 : this system uses ETED with manually assigned costs. The costs for single nodes are the same for the ZS-TED 1 experiment and the costs for subtrees are half the sum of the costs of their parts.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "ETED 2 : this system uses ETED with the intuition-based costs for single nodes given in Figure 3 (column A) and the costs for subtrees given in Figure 3 (column B). ETED+ABC: this system uses the ABC algorithm to estimate the costs of edit single operations and threshold(s) for ETED. For binary decision output, the chromosome is {cost of deleting a node, cost of inserting a node, cost of exchanging a node, multiplier for the sum of the costs of the deletions in a deleted subtree, multiplier for the sum of the costs of the insertions in an inserted subtree, multiplier for the sum of the costs of the exchanges in an exchanged subtree, threshold}. For three-way decisions the chromosome also contains the second threshold. For both cases the fitness is as for ZS-TED+GA. We do not include GA results for ETED, as extensive comparison of the standard GA algorithm and ABC on the ZS-TED experiments shows that ABC consistently produces better results for the same initial seeds and the same number of iterations. The BoW algorithm and the basic string-edit algorithm are supplemented by the first two of the three procedures listed below and the others by all three, to ensure that we get the best possible performance at each stage:",
"cite_spans": [],
"ref_spans": [
{
"start": 88,
"end": 94,
"text": "Figure",
"ref_id": null
},
{
"start": 144,
"end": 152,
"text": "Figure 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "\u2022 use AWN, OpenOffice Arabic dictionary and others as a lexical resource in order to take account of synonymy, antonym and hyponymy relations when comparing two words and when calculating the cost of an edit;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "\u2022 take into consideration the POS tag when comparing two similar words (i.e. they should have the same POS tag);",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "\u2022 use a list of stopwords that contains some of the commonest Arabic words, which are treated specially when comparing words (e.g. by using different edit costs for them in distance-based approaches).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Syntactic similarity approaches",
"sec_num": null
},
{
"text": "We carried out experiments using the approaches above with two types of decisions as below.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Results",
"sec_num": "4.2"
},
{
"text": "Simple binary decision ('yes' and 'no'): T entails H when the cost of matching is less (more in case of bag-of-words) than a threshold. The results of this experiments, in terms of precision (P), recall (R) and f-score (F) for 'yes' class and accuracy (Acc.), are shown in Table 1 . ETED shows a substantial improvement over bag-of-words and Levenshtein distance (around 19% in f-score and 6% in total accuracy) and over ZS-TED (around 2% in f-score and 2% in total accuracy). Although we are primarily interested in Arabic, we have carried out parallel sets of experiments on the English RTE2 parsed testset, 1 using the Princeton WordNet (PWN) as a lexical resource, with the input text converted to dependency trees using Minipar (Lin, 1998) . The pattern in Table 1 for English is similar to that for Arabic. ZS-TED is better than bag-of-words, ETED is a further improvement over ZS-TED. -way decision ('yes,' 'unknown' and 'no' (not 'contradicts' ) ): for this task we use two thresholds, one to trigger a positive answer if the cost of matching is lower than the lower threshold (exceeds the higher one for the bag-ofwords algorithm) and the other to trigger a negative answer if the cost of matching exceeds the higher one (mutatis mutandis for bag-of-words). Otherwise, the result will be 'unknown.' The reason for making a three-way decision is to drive systems to make more precise distinctions. There is a difference between knowing that H does not Cost (A) Single node (B) Subtree (more than one node) Delete if X is a stop word =5, 0 else =7 Insert if Y is a stop word =5, double the sum of the costs of its parts else =100 Exchange if X subsumes Y =0, if a subtree S1 is identical to a subtree S2=0 if X is a stop word =5, else half the sum of the costs of its parts if Y subsumes or contradicts X=100 else =50 Figure 3 : Intuition-based edit operation costs for the systems ZS-TED2 and ETED1 (X in T , Y in H).",
"cite_spans": [
{
"start": 733,
"end": 744,
"text": "(Lin, 1998)",
"ref_id": "BIBREF21"
}
],
"ref_spans": [
{
"start": 273,
"end": 280,
"text": "Table 1",
"ref_id": null
},
{
"start": 762,
"end": 769,
"text": "Table 1",
"ref_id": null
},
{
"start": 892,
"end": 952,
"text": "-way decision ('yes,' 'unknown' and 'no' (not 'contradicts'",
"ref_id": null
},
{
"start": 1826,
"end": 1834,
"text": "Figure 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Results",
"sec_num": "4.2"
},
{
"text": "Three Table 1 : Comparison between ETED, simple bag-of-words, Levenshtein distance and ZS-TED.",
"cite_spans": [],
"ref_spans": [
{
"start": 6,
"end": 13,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Binary decision",
"sec_num": null
},
{
"text": "entail T and not knowing whether it does or not. Note that answering 'no' here means \"I believe that H does not entail T \", not \"I believe that H contradicts T .\" The results of this experiment, in terms of precision, recall and f-score for 'yes' class, are shown in Table 1 . Again, ETED shows a worthwhile improvement bag-of-words and Levenshtein distance (around 6% in f-score) and over ZS-TED (around 4% in f-score).",
"cite_spans": [],
"ref_spans": [
{
"start": 267,
"end": 274,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Binary decision",
"sec_num": null
},
{
"text": "We have described an extended version of tree edit distance (TED) that allows operations (i.e. delete, insert and exchange) both on single nodes and on subtrees. The extended TED with subtree operations, ETED, is more effective and flexible than the ZS-TED, especially for applications that pay attention to relations among nodes (e.g. in linguistic trees, deleting a modifier subtree should be cheaper than the sum of deleting its components individually).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Summary",
"sec_num": "5"
},
{
"text": "We have also investigated the use of different optimisation algorithms, and have shown that using these produces better performance than setting the costs of edit operations by hand, and that using the ABC algorithm produces better results for the same amount of effort as traditional GAs.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Summary",
"sec_num": "5"
},
{
"text": "The current findings, while preliminary, are quite encouraging. The fact that the results on our original testset, particularly the improvement in f-score, were replicated for a testset where we had no control over the parser that was used to produce dependency trees from the T-H pairs provides some evidence for the robustness of the approach. We anticipate that in both cases having a more accurate parser (our parser for Arabic attains around 85% accuracy on the PATB, Minipar is reported to attain about 80% on the Suzanne corpus) would improve the performance of both ZS-TED and ETED.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Summary",
"sec_num": "5"
},
{
"text": "http://u.cs.biu.ac.il/~nlp/RTE2/Datasets/RTE-2\\ %20Preprocessed\\%20Datasets.html",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "Maytham Alabbas owes his deepest gratitude to Iraqi Ministry of Higher Education and Scientific Research for financial support in his PhD study. Allan Ramsay's contribution to this work was partially supported by the Qatar National Research Fund (grant NPRP 09 -046 -6 -001).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgements",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "A modified artificial bee colony algorithm for real-parameter optimization",
"authors": [
{
"first": "B",
"middle": [],
"last": "Akay",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Karaboga",
"suffix": ""
}
],
"year": 2012,
"venue": "Information Sciences",
"volume": "192",
"issue": "0",
"pages": "120--142",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "B. Akay and D. Karaboga. 2012. A modified artificial bee colony algorithm for real-parameter optimiza- tion. Information Sciences, 192(0):120 -142.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Evaluation of combining data-driven dependency parsers for Arabic",
"authors": [
{
"first": "M",
"middle": [],
"last": "Alabbas",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Ramsay",
"suffix": ""
}
],
"year": 2011,
"venue": "Proceeding of 5th Language & Technology Conference: Human Language Technologies (LTC'11)",
"volume": "",
"issue": "",
"pages": "546--550",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Alabbas and A. Ramsay. 2011. Evaluation of com- bining data-driven dependency parsers for Arabic. In Proceeding of 5th Language & Technology Con- ference: Human Language Technologies (LTC'11), pages 546-550, Pozna\u0144, Poland.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Improved POStagging for Arabic by combining diverse taggers",
"authors": [
{
"first": "M",
"middle": [],
"last": "Alabbas",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Ramsay",
"suffix": ""
}
],
"year": 2012,
"venue": "Artificial Intelligence Applications and Innovations (AIAI)",
"volume": "381",
"issue": "",
"pages": "107--116",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Alabbas and A. Ramsay. 2012. Improved POS- tagging for Arabic by combining diverse taggers. In L. Iliadis, I. Maglogiannis, and H. Papadopoulos, editors, Artificial Intelligence Applications and In- novations (AIAI), volume 381 of IFIP Advances in Information and Communication Technology, pages 107-116. Springer Berlin Heidelberg, Halkidiki, Thessaloniki, Greece.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "ArbTE: Arabic textual entailment",
"authors": [
{
"first": "M",
"middle": [],
"last": "Alabbas",
"suffix": ""
}
],
"year": 2011,
"venue": "Proceedings of the Second Student Research Workshop associated with RANLP 2011",
"volume": "",
"issue": "",
"pages": "48--53",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Alabbas. 2011. ArbTE: Arabic textual en- tailment. In Proceedings of the Second Student Research Workshop associated with RANLP 2011, pages 48-53, Hissar, Bulgaria. RANLP 2011 Organ- ising Committee.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "A dataset for Arabic Textual Entailment",
"authors": [
{
"first": "M",
"middle": [],
"last": "Alabbas",
"suffix": ""
}
],
"year": 2013,
"venue": "Proceedings of the Second Student Research Workshop associated with RANLP 2013",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Alabbas. 2013. A dataset for Arabic Textual Entailment. In Proceedings of the Second Student Research Workshop associated with RANLP 2013, Hissar, Bulgaria. RANLP 2013 Organising Commit- tee.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Semantic inference at the lexical-syntactic level",
"authors": [
{
"first": "R",
"middle": [],
"last": "Bar-Haim",
"suffix": ""
},
{
"first": "I",
"middle": [],
"last": "Dagan",
"suffix": ""
},
{
"first": "I",
"middle": [],
"last": "Greental",
"suffix": ""
},
{
"first": "E",
"middle": [],
"last": "Shnarch",
"suffix": ""
}
],
"year": 2007,
"venue": "Proceedings of 22nd AAAI Conference on Artificial Intelligence (AAAI-07)",
"volume": "",
"issue": "",
"pages": "871--876",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "R. Bar-Haim, I. Dagan, I. Greental, and E. Shnarch. 2007. Semantic inference at the lexical-syntactic level. In Proceedings of 22nd AAAI Conference on Artificial Intelligence (AAAI-07), pages 871-876, Vancouver, British Columbia, Canada. The AAAI Press.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "A survey on tree edit distance and related problems",
"authors": [
{
"first": "P",
"middle": [],
"last": "Bille",
"suffix": ""
}
],
"year": 2005,
"venue": "Theoretical Computer Science",
"volume": "337",
"issue": "1-3",
"pages": "217--239",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "P. Bille. 2005. A survey on tree edit distance and related problems. Theoretical Computer Science, 337(1-3):217-239.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Introducing the Arabic WordNet project",
"authors": [
{
"first": "W",
"middle": [],
"last": "Black",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Elkateb",
"suffix": ""
},
{
"first": "H",
"middle": [],
"last": "Rodriguez",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Alkhalifa",
"suffix": ""
},
{
"first": "P",
"middle": [],
"last": "Vossen",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Pease",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Fellbaum",
"suffix": ""
}
],
"year": 2006,
"venue": "Proceedings of the 3rd International WordNet Conference (GWC-06)",
"volume": "",
"issue": "",
"pages": "295--299",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "W. Black, S. Elkateb, H. Rodriguez, M. Alkhalifa, P. Vossen, A. Pease, and C. Fellbaum. 2006. Intro- ducing the Arabic WordNet project. In Proceedings of the 3rd International WordNet Conference (GWC- 06), pages 295-299, Jeju Island, Korea.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "The PASCAL recognising textual entailment challenge",
"authors": [
{
"first": "I",
"middle": [],
"last": "Dagan",
"suffix": ""
},
{
"first": "O",
"middle": [],
"last": "Glickman",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Magnini",
"suffix": ""
}
],
"year": 2005,
"venue": "Proceedings of the 1st PASCAL Recognising Textual Entailment Challenges",
"volume": "",
"issue": "",
"pages": "1--8",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "I. Dagan, O. Glickman, and B. Magnini. 2005. The PASCAL recognising textual entailment challenge. In Proceedings of the 1st PASCAL Recognising Tex- tual Entailment Challenges, pages 1-8, Southamp- ton, UK.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "An optimal decomposition algorithm for tree edit distance",
"authors": [
{
"first": "E",
"middle": [],
"last": "Demaine",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Mozes",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Rossman",
"suffix": ""
},
{
"first": "O",
"middle": [],
"last": "Weimann",
"suffix": ""
}
],
"year": 2009,
"venue": "ACM Transactions on Algorithms (TALG)",
"volume": "6",
"issue": "1",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "E. Demaine, S. Mozes, B. Rossman, and O. Weimann. 2009. An optimal decomposition algorithm for tree edit distance. ACM Transactions on Algorithms (TALG), 6(1):2:1-2:19.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "Second generation tools (AMIRA 2.0): fast and robust tokenization, POS tagging, and base phrase chunking",
"authors": [
{
"first": "M",
"middle": [],
"last": "Diab",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 2nd International Conference on Arabic Language Resources and Tools",
"volume": "",
"issue": "",
"pages": "285--288",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Diab. 2009. Second generation tools (AMIRA 2.0): fast and robust tokenization, POS tagging, and base phrase chunking. In Proceedings of the 2nd International Conference on Arabic Language Re- sources and Tools, pages 285-288, Cairo, Eygpt. The MEDAR Consortium.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "MADA+TOKAN: a toolkit for Arabic tokenization, diacritization, morphological disambiguation, POS tagging, stemming and lemmatization",
"authors": [
{
"first": "N",
"middle": [],
"last": "Habash",
"suffix": ""
},
{
"first": "O",
"middle": [],
"last": "Rambow",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Roth",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 2nd International Conference on Arabic Language Resources and Tools",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "N. Habash, O. Rambow, and R. Roth. 2009. MADA+TOKAN: a toolkit for Arabic tokenization, diacritization, morphological disambiguation, POS tagging, stemming and lemmatization. In Proceed- ings of the 2nd International Conference on Arabic Language Resources and Tools, Cairo, Eygpt. The MEDAR Consortium.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Inferring textual entailment with a probabilistically sound calculus",
"authors": [
{
"first": "S",
"middle": [],
"last": "Harmeling",
"suffix": ""
}
],
"year": 2009,
"venue": "Natural Language Engineering",
"volume": "15",
"issue": "4",
"pages": "459--477",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "S. Harmeling. 2009. Inferring textual entailment with a probabilistically sound calculus. Natural Lan- guage Engineering, 15(4):459-477.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Tree edit models for recognizing textual entailments, paraphrases, and answers to questions",
"authors": [
{
"first": "M",
"middle": [],
"last": "Heilman",
"suffix": ""
},
{
"first": "N",
"middle": [],
"last": "Smith",
"suffix": ""
}
],
"year": 2010,
"venue": "Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics",
"volume": "",
"issue": "",
"pages": "1011--1019",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Heilman and N. Smith. 2010. Tree edit models for recognizing textual entailments, paraphrases, and answers to questions. In Human Language Tech- nologies: The 2010 Annual Conference of the North American Chapter of the Association for Compu- tational Linguistics, pages 1011-1019, Los Ange- les, California, USA. Association for Computational Linguistics.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Textual entailment recognition based on dependency analysis and WordNet",
"authors": [
{
"first": "J",
"middle": [],
"last": "Herrera",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Pe\u00f1as",
"suffix": ""
},
{
"first": "F",
"middle": [],
"last": "Verdejo",
"suffix": ""
}
],
"year": 2005,
"venue": "Proceedings of PASCAL Workshop on Recognizing Textual Entailment",
"volume": "",
"issue": "",
"pages": "21--24",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. Herrera, A. Pe\u00f1as, and F. Verdejo. 2005. Textual entailment recognition based on dependency analy- sis and WordNet. In Proceedings of PASCAL Work- shop on Recognizing Textual Entailment, pages 21- 24, Southampton, UK.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "UNED at PASCAL RTE-2 challenge",
"authors": [
{
"first": "J",
"middle": [],
"last": "Herrera",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Pe\u00f1as",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Rodrigo",
"suffix": ""
},
{
"first": "F",
"middle": [],
"last": "Verdejo",
"suffix": ""
}
],
"year": 2006,
"venue": "Proceedings of the 2nd PASCAL Challenges Workshop on Recognising Textual Entailment",
"volume": "",
"issue": "",
"pages": "38--43",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. Herrera, A. Pe\u00f1as, A. Rodrigo, and F. Verdejo. 2006. UNED at PASCAL RTE-2 challenge. In Proceedings of the 2nd PASCAL Challenges Work- shop on Recognising Textual Entailment, pages 38- 43, Venice, Italy.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Tree matching problems with applications to structured text databases",
"authors": [
{
"first": "P",
"middle": [],
"last": "Kilpel\u00e4inen",
"suffix": ""
}
],
"year": 1992,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "P. Kilpel\u00e4inen. 1992. Tree matching problems with applications to structured text databases. Technical Report A-1992-6, Department of Computer Science, University of Helsinki, Helsinki, Finland.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "A tree-edit-distance algorithm for comparing simple, closed shapes",
"authors": [
{
"first": "P",
"middle": [],
"last": "Klein",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Tirthapura",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Sharvit",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Kimia",
"suffix": ""
}
],
"year": 2000,
"venue": "Proceedings of the eleventh annual ACM-SIAM symposium on Discrete algorithms",
"volume": "",
"issue": "",
"pages": "696--704",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "P. Klein, S. Tirthapura, D. Sharvit, and B. Kimia. 2000. A tree-edit-distance algorithm for comparing sim- ple, closed shapes. In Proceedings of the eleventh annual ACM-SIAM symposium on Discrete algo- rithms, pages 696-704. Society for Industrial and Applied Mathematics.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "Computing the edit-distance between unrooted ordered trees",
"authors": [
{
"first": "P",
"middle": [],
"last": "Klein",
"suffix": ""
}
],
"year": 1998,
"venue": "Proceedings of the 6th Annual European Symposium on Algorithms (ESA '98)",
"volume": "",
"issue": "",
"pages": "91--102",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "P. Klein. 1998. Computing the edit-distance between unrooted ordered trees. In Proceedings of the 6th Annual European Symposium on Algorithms (ESA '98), pages 91-102, Venice, Italy. Springer-Verlag.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "Recognizing textual entailment with tree edit distance algorithms",
"authors": [
{
"first": "M",
"middle": [],
"last": "Kouylekov",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Magnini",
"suffix": ""
}
],
"year": 2005,
"venue": "Proceedings of the 1st PASCAL Recognising Textual Entailment Challenge",
"volume": "",
"issue": "",
"pages": "17--20",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Kouylekov and B. Magnini. 2005. Recognizing textual entailment with tree edit distance algorithms. In Proceedings of the 1st PASCAL Recognising Tex- tual Entailment Challenge, pages 17-20, Southamp- ton, UK.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Dependency Parsing",
"authors": [
{
"first": "S",
"middle": [],
"last": "K\u00fcbler",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Mcdonald",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Nivre",
"suffix": ""
}
],
"year": 2009,
"venue": "Synthesis Lectures on Human Language Technologies. Morgan & Claypool Publishers",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "S. K\u00fcbler, R. McDonald, and J. Nivre. 2009. Depen- dency Parsing. Synthesis Lectures on Human Lan- guage Technologies. Morgan & Claypool Publish- ers.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "Dependency-based evaluation of MINI-PAR",
"authors": [
{
"first": "D",
"middle": [],
"last": "Lin",
"suffix": ""
}
],
"year": 1998,
"venue": "LREC'98: Workshop on the Evaluation of Parsing systems",
"volume": "",
"issue": "",
"pages": "317--330",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "D. Lin. 1998. Dependency-based evaluation of MINI- PAR. In LREC'98: Workshop on the Evaluation of Parsing systems, pages 317-330, Granada, Spain.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Normalized alignment of dependency trees for detecting textual entailment",
"authors": [
{
"first": "E",
"middle": [],
"last": "Marsi",
"suffix": ""
},
{
"first": "E",
"middle": [],
"last": "Krahmer",
"suffix": ""
},
{
"first": "W",
"middle": [],
"last": "Bosma",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Theune",
"suffix": ""
}
],
"year": 2006,
"venue": "Proceedings of the 2nd PASCAL Challenges Workshop on Recognising Textual Entailment",
"volume": "",
"issue": "",
"pages": "56--61",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "E. Marsi, E. Krahmer, W. Bosma, and M. Theune. 2006. Normalized alignment of dependency trees for detecting textual entailment. In Proceedings of the 2nd PASCAL Challenges Workshop on Recognis- ing Textual Entailment, pages 56-61, Venice, Italy.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "Multilingual dependency parsing with a two-stage discriminative parser",
"authors": [
{
"first": "R",
"middle": [],
"last": "Mcdonald",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Lerman",
"suffix": ""
},
{
"first": "F",
"middle": [],
"last": "Pereira",
"suffix": ""
}
],
"year": 2006,
"venue": "10th Conference on Computational Natural Language Learning (CoNLL-X)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "R. McDonald, K. Lerman, and F. Pereira. 2006. Mul- tilingual dependency parsing with a two-stage dis- criminative parser. In 10th Conference on Computa- tional Natural Language Learning (CoNLL-X), New York, USA.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "Optimizing textual entailment recognition using particle swarm optimization",
"authors": [
{
"first": "Y",
"middle": [],
"last": "Mehdad",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Magnini",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 2009 Workshop on Applied Textual Inference (TextInfer '09)",
"volume": "",
"issue": "",
"pages": "36--43",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Y. Mehdad and B. Magnini. 2009. Optimizing tex- tual entailment recognition using particle swarm op- timization. In Proceedings of the 2009 Workshop on Applied Textual Inference (TextInfer '09), pages 36-43, Suntec, Singapore. Association for Compu- tational Linguistics.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Alignment of shared forests for bilingual corpora",
"authors": [
{
"first": "A",
"middle": [],
"last": "Meyers",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Yangarber",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Grishman",
"suffix": ""
}
],
"year": 1996,
"venue": "Proceedings of the 16th Conference on Computational Linguistics",
"volume": "1",
"issue": "",
"pages": "460--465",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "A. Meyers, R. Yangarber, and R. Grishman. 1996. Alignment of shared forests for bilingual corpora. In Proceedings of the 16th Conference on Com- putational Linguistics, volume 1, pages 460-465, Copenhagen, Denmark. Association for Computa- tional Linguistics.",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Malt-Parser: a language-independent system for datadriven dependency parsing",
"authors": [
{
"first": "J",
"middle": [],
"last": "Nivre",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Hall",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Nilsson",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Chanev",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Eryigit",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "K\u00fcbler",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Marinov",
"suffix": ""
},
{
"first": "E",
"middle": [],
"last": "Marsi",
"suffix": ""
}
],
"year": 2007,
"venue": "Natural Language Engineering",
"volume": "13",
"issue": "02",
"pages": "95--135",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. Nivre, J. Hall, J. Nilsson, A. Chanev, G. Eryigit, S. K\u00fcbler, S. Marinov, and E. Marsi. 2007. Malt- Parser: a language-independent system for data- driven dependency parsing. Natural Language En- gineering, 13(02):95-135.",
"links": null
},
"BIBREF27": {
"ref_id": "b27",
"title": "RTED: a robust algorithm for the tree edit distance",
"authors": [
{
"first": "M",
"middle": [],
"last": "Pawlik",
"suffix": ""
},
{
"first": "N",
"middle": [],
"last": "Augsten",
"suffix": ""
}
],
"year": 2011,
"venue": "Proceedings of the VLDB Endowment",
"volume": "5",
"issue": "",
"pages": "334--345",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Pawlik and N. Augsten. 2011. RTED: a robust algorithm for the tree edit distance. Proceedings of the VLDB Endowment, 5(4):334-345.",
"links": null
},
"BIBREF28": {
"ref_id": "b28",
"title": "Natural language inference via dependency tree mapping: An application to question answering",
"authors": [
{
"first": "V",
"middle": [],
"last": "Punyakanok",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Roth",
"suffix": ""
},
{
"first": "W",
"middle": [],
"last": "Yih",
"suffix": ""
}
],
"year": 2004,
"venue": "Computational Linguistics",
"volume": "6",
"issue": "",
"pages": "1--10",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "V. Punyakanok, D. Roth, and W. Yih. 2004. Nat- ural language inference via dependency tree map- ping: An application to question answering. Com- putational Linguistics, 6:1-10.",
"links": null
},
"BIBREF29": {
"ref_id": "b29",
"title": "Bootstrapping a lexicon-free tagger for Arabic",
"authors": [
{
"first": "A",
"middle": [],
"last": "Ramsay",
"suffix": ""
},
{
"first": "Y",
"middle": [],
"last": "Sabtan",
"suffix": ""
}
],
"year": 2009,
"venue": "Proceedings of the 9th Conference on Language Engineering",
"volume": "",
"issue": "",
"pages": "202--215",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "A. Ramsay and Y. Sabtan. 2009. Bootstrapping a lexicon-free tagger for Arabic. In Proceedings of the 9th Conference on Language Engineering (ES- OLEC'2009), pages 202-215, Cairo, Egypt.",
"links": null
},
"BIBREF30": {
"ref_id": "b30",
"title": "Particle Swarm Optimization",
"authors": [
{
"first": "J",
"middle": [],
"last": "Russell",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Cohn",
"suffix": ""
}
],
"year": 2012,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. Russell and R. Cohn. 2012. Particle Swarm Opti- mization. Book on Demand Ltd.",
"links": null
},
"BIBREF31": {
"ref_id": "b31",
"title": "A confidence model for syntactically-motivated entailment proofs",
"authors": [
{
"first": "A",
"middle": [],
"last": "Stern",
"suffix": ""
},
{
"first": "I",
"middle": [],
"last": "Dagan",
"suffix": ""
}
],
"year": 2011,
"venue": "Proceedings of Recent Advances in Natural Language Processing",
"volume": "",
"issue": "",
"pages": "455--462",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "A. Stern and I. Dagan. 2011. A confidence model for syntactically-motivated entailment proofs. In Pro- ceedings of Recent Advances in Natural Language Processing (RANLP 2011), pages 455-462, Hissar, Bulgaria. RANLP 2011 Organising Committee.",
"links": null
},
"BIBREF32": {
"ref_id": "b32",
"title": "Efficient search for transformation-based inference",
"authors": [
{
"first": "A",
"middle": [],
"last": "Stern",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Stern",
"suffix": ""
},
{
"first": "I",
"middle": [],
"last": "Dagan",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Felner",
"suffix": ""
}
],
"year": 2012,
"venue": "Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics",
"volume": "1",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "A. Stern, R. Stern, I. Dagan, and A. Felner. 2012. Ef- ficient search for transformation-based inference. In Proceedings of the 50th Annual Meeting of the As- sociation for Computational Linguistics, volume 1, Jeju Island, Korea. The Association for Computer Linguistics.",
"links": null
},
"BIBREF33": {
"ref_id": "b33",
"title": "Probabilistic treeedit models with structured latent variables for textual entailment and question answering",
"authors": [
{
"first": "M",
"middle": [],
"last": "Wang",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Manning",
"suffix": ""
}
],
"year": 2010,
"venue": "Proceedings of the 23rd International Conference on Computational Linguistics",
"volume": "",
"issue": "",
"pages": "1164--1172",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "M. Wang and C. Manning. 2010. Probabilistic tree- edit models with structured latent variables for tex- tual entailment and question answering. In Proceed- ings of the 23rd International Conference on Com- putational Linguistics (Coling 2010), pages 1164- 1172, Beijing, China. Association for Computa- tional Linguistics, Tsinghua University Press.",
"links": null
},
"BIBREF34": {
"ref_id": "b34",
"title": "Simple fast algorithms for the editing distance between trees and related problems",
"authors": [
{
"first": "K",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Shasha",
"suffix": ""
}
],
"year": 1989,
"venue": "SIAM Journal of Computing",
"volume": "18",
"issue": "6",
"pages": "1245--1262",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "K. Zhang and D. Shasha. 1989. Simple fast algo- rithms for the editing distance between trees and related problems. SIAM Journal of Computing, 18(6):1245-1262.",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"type_str": "figure",
"num": null,
"uris": null,
"text": ". For each mapping figure the insertion, deletion, matching and exchange operations are shown with single, double, single dashed and double dashed outline respectively. The matching nodes (or subtrees) are linked with dashed arrows. Standard TED, mapping between T 1 and T 2 ."
},
"FIGREF1": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Extended TED with subtree operations, mapping between T 1 and T 2 ."
},
"FIGREF2": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "size of population. D number of optimisation parameters. xij solution i,j, i = 1 ...SN, j = 1 ...D 1: Initialise the population of solutions xi,j, i = 1...SN, j = 1...D, triali = 0vij for the employed bees (using vij = xij + \u03c6ij(xij \u2212 x kj ), where k \u2208 {1, ..., SN} and \u03c6ij is a random number between [-1,1]) and evaluate them; 6:"
}
}
}
} |