File size: 147,630 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 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 | {
"paper_id": "2020",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T13:04:59.460272Z"
},
"title": "Towards Automatic Thesaurus Construction and Enrichment",
"authors": [
{
"first": "Claudia",
"middle": [],
"last": "Lanza",
"suffix": "",
"affiliation": {},
"email": "c.lanza@dimes.unical.it"
},
{
"first": "Amir",
"middle": [],
"last": "Hazem",
"suffix": "",
"affiliation": {
"laboratory": "LS2N -UMR CNRS 6004",
"institution": "Universit\u00e9 de Nantes",
"location": {
"country": "France"
}
},
"email": "amir.hazem@ls2n.fr"
},
{
"first": "B\u00e9atrice",
"middle": [],
"last": "Daille",
"suffix": "",
"affiliation": {
"laboratory": "LS2N -UMR CNRS 6004",
"institution": "Universit\u00e9 de Nantes",
"location": {
"country": "France"
}
},
"email": "beatrice.daille@ls2n.fr"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "Thesaurus construction with minimum human efforts often relies on automatic methods to discover terms and their relations. Hence, the quality of a thesaurus heavily depends on the chosen methodologies for: (i) building its content (terminology extraction task) and (ii) designing its structure (semantic similarity task). The performance of the existing methods on automatic thesaurus construction is still less accurate than the handcrafted ones of which is important to highlight the drawbacks to let new strategies build more accurate thesauri models. In this paper, we will provide a systematic analysis of existing methods for both tasks and discuss their feasibility based on an Italian Cybersecurity corpus. In particular, we will provide a detailed analysis on how the semantic relationships network of a thesaurus can be automatically built, and investigate the ways to enrich the terminological scope of a thesaurus by taking into account the information contained in external domain-oriented semantic sets.",
"pdf_parse": {
"paper_id": "2020",
"_pdf_hash": "",
"abstract": [
{
"text": "Thesaurus construction with minimum human efforts often relies on automatic methods to discover terms and their relations. Hence, the quality of a thesaurus heavily depends on the chosen methodologies for: (i) building its content (terminology extraction task) and (ii) designing its structure (semantic similarity task). The performance of the existing methods on automatic thesaurus construction is still less accurate than the handcrafted ones of which is important to highlight the drawbacks to let new strategies build more accurate thesauri models. In this paper, we will provide a systematic analysis of existing methods for both tasks and discuss their feasibility based on an Italian Cybersecurity corpus. In particular, we will provide a detailed analysis on how the semantic relationships network of a thesaurus can be automatically built, and investigate the ways to enrich the terminological scope of a thesaurus by taking into account the information contained in external domain-oriented semantic sets.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "In computational linguistics and terminology, a thesaurus is often used to represent the knowledge of a specific domain of study as a controlled vocabulary. This paper aims at presenting an analysis of the best performing NLP approaches, i.e., patterns configuration, semantic similarity, morphosyntactic variation given by term extractors, in enhancing a semantic structure of an existing Italian thesaurus about the technical domain of Cybersecurity. Constructing thesauri by carrying out minimum handcrafted activities is currently highly demanded (Azevedo et al., 2015) . Hence, several methods to automatically build and maintain a thesaurus have been proposed so far (G\u00fcntzer et al., 1989; Morin and Jacquemin, 1999; Yang and Powers, 2008a; Schandl and Blumauer, 2010) . However, the quality of automatically generated thesauri tends to be rather weaker in their content and structure with respect to the conventional handcrafted ones (Ryan, 2014) . To guarantee the currency of a thesaurus (Batini et al., 2009) it is crucial to whether improve existing methods or to develop new efficient techniques for discovering terms and their relations. On the perspective of using existing NLP tools for constructing a thesaurus, choosing the most appropriate ones is not an easy task since the performance varies depending on the domain (Nielsen, 2001) , the supported languages, the applied strategies, etc. Selecting a highly performing NLP procedure to build on a knowledge representation resource does also contemplate maintenance and enrichment phases aimed at empowering the application usages of these semantic sources. This work aims at presenting an analysis of which of the NLP approaches, i.e., patterns configuration, semantic similarity, morpho-syntactic variation given by term extractors, could be considered the best performing in enhancing a semantic structure of an existing Italian thesaurus about the technical domain of Cybersecurity. The paper starts firstly from a description of how the current thesaurus has been constructed (Broughton, 2008) , following the rules included in the main reference standards for building thesauri (ISO/TC 46/SC 9 2011 and 2013), and of the source cor-pora composition from which the thesaurus construction has taken its basis. In detail, the paper is organized as follows: Section 2 presents a state-of-the-art on the main works about the construction of terminological knowledge bases, as well as on those that dealt with the semantic relations discovering approaches, such as, the distributional similarity ones. Section 3 describes the former configuration of the handcrafted thesaurus for Cybersecurity and of the source corpus used to build the controlled vocabulary on the Cybersecurity domain, i.e., the Italian corpus made up of legislation and domain-oriented magazines. Section 4 provides an outline of the data sets, i.e., a ranked summary of the terminological lists, including the ones considered as the main gold standards to which rely on; in this part a set of representative examples for each existing relation, which has been extracted from the draft thesaurus to use as data meant to be ameliorated, is given. Section 5 to 7 describe the methods used to automatize the hierarchical, associative and synonymous configuration of the Italian Cybersecurity thesaurus along with their experiments and results. Section 8 combines the results to determine which approach is the best performing with respect to the desired thesaurus output to achieve. Finally, Section 8 presents the conclusion.",
"cite_spans": [
{
"start": 551,
"end": 573,
"text": "(Azevedo et al., 2015)",
"ref_id": "BIBREF2"
},
{
"start": 673,
"end": 695,
"text": "(G\u00fcntzer et al., 1989;",
"ref_id": "BIBREF22"
},
{
"start": 696,
"end": 722,
"text": "Morin and Jacquemin, 1999;",
"ref_id": "BIBREF38"
},
{
"start": 723,
"end": 746,
"text": "Yang and Powers, 2008a;",
"ref_id": "BIBREF49"
},
{
"start": 747,
"end": 774,
"text": "Schandl and Blumauer, 2010)",
"ref_id": "BIBREF44"
},
{
"start": 941,
"end": 953,
"text": "(Ryan, 2014)",
"ref_id": "BIBREF42"
},
{
"start": 997,
"end": 1018,
"text": "(Batini et al., 2009)",
"ref_id": "BIBREF6"
},
{
"start": 1336,
"end": 1351,
"text": "(Nielsen, 2001)",
"ref_id": "BIBREF40"
},
{
"start": 2049,
"end": 2066,
"text": "(Broughton, 2008)",
"ref_id": "BIBREF11"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1."
},
{
"text": "The main purpose presented in this paper is to guarantee a higher-quality management of the Italian Cybersecurity thesaurus' domain-oriented terminology. In particular, this paper explores which could be considered the best performing NLP tool among a plethora of selected ones to be used in order to empower an existing thesaurus of a highly technical domain, the Cybersecurity one. The source language of this semantic resource is the Italian, and the methods pursued to provide reliable candidate terms structures, meant to be included in the thesaurus, are based on sophisticated terminological extractor tools. With the objective of carrying out a study on how to automatically generate the semantic networking systems proper to theauri, these terms extraction software represent the basis from which to be-gin the non-manually construction of a thesaurus outline. Specifically, the approaches undertaken are the following:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Objectives",
"sec_num": "2."
},
{
"text": "1. Pattern based system: the causative patterns aim at enhancing the associative relationship proper to thesauri configuration;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Objectives",
"sec_num": "2."
},
{
"text": "2. Variants recognition: semantic variation is useful to detect hierarchical and associative sets;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Objectives",
"sec_num": "2."
},
{
"text": "3. Distributional analysis: this procedural methodology helps in identifying the synonymy connection.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Objectives",
"sec_num": "2."
},
{
"text": "Automatically constructing a thesaurus aims at obtaining, as output, an improved knowledge organization system on the Cybersecurity area of study from a semantic correlation construction point of view. This system should provide an advanced hierarchical structuring that is meant to overcome a current thesaurus outline, as well as the associative and equivalence terms organization. Indeed, as described in the following sections, the handcrafted thesaurus categorization sometimes proves to be either subjective and not completely explicit in representing associations among domain-specific terms.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Objectives",
"sec_num": "2."
},
{
"text": "A thesaurus can be considered as a controlled system that organizes the knowledge of a specific domain of study through a network of semantic relations linked to the hierarchy, synonymy and association structures (Broughton, 2008) . Terms included in the thesauri have to keep a unambiguous value, as affirmed in the standard NISO TR-06-2017, Issues in Vocabulary Management: \"Controlled vocabulary: A list of terms that have been enumerated explicitly. This list is controlled by and is available from a controlled vocabulary registration authority. All terms in a controlled vocabulary must have an unambiguous, nonredundant definition\". Constructing an efficient terminological system usually implies the acquisition of domainoriented information from texts, specifically those that can provide semantic knowledge density and granularity about the lexicon that is meant to be represented (Barri\u00e8re, 2006) . These structures are in literature known as TKBs (Terminological Knowledge Bases) (Condamines, 2018) , and, indeed, they support the modalities of systematizing the specialized knowledge by merging the skills proper to linguistics and knowledge engineering. The ways in which the candidate terms are extracted from a specific domainoriented corpus (Loginova Clouet et al., 2012) usually follow text pre-processing procedures and extraction of single and multi-word units (Daille and Hazem, 2014) from texts filtered out by frequency measures, then they can undergo a phase of variation recognition (Weller et al., 2011) and other statistical calculations to determine the specificity, accuracy, similarity in the texts from which they come from (Cabr\u00e9 et al., 2001) . The reason why the domain-oriented terms are called 'candidates' (Condamines, 2018) is linked to the fact that in the terminologists' activity the need of experts' validation is frequently required, this because just the subjective selection by terminologists might not be exhaustive and fully consistent with the domain expertise (ISO/TC 46/SC 9 2013). Thesauri's realization is commonly connoted by a manual semantic work that assumes a terminologists' activity in selecting terms from a list of candidate ones, extracted, in turn, from a reference corpus (Condamines, 2007) and, consequently, arranging them in a structure that follows the guidelines given by ISO standards for constructing thesauri (ISO/TC 46/SC 9 2011 and 2013) which aim at normalizing the information meant to be shared by a community of users. For the seek of gaining time to terminologists in defining thesauri's structure (Rennesson et al., 2020) , their construction phases are supported by using computer engineering techniques and followed by an evaluation phase that sees experts of the domain involved in the decisionmaking process about the insertion of the terms in the semantic resource. Even though, a process of appropriateness' check by experts isn't entirely suitable to demonstrate that the TKBs comply with the specialized corpus knowledge flow. Hence, together with certain groups of experts' supervision, other tools support the accuracy validation, i.e., the gold standards (Barri\u00e8re, 2006) . This task is meant to give results on the way terms that have been selected to be part of a semantic resource -designed to represent a specialized language -can be aligned with others included in reference texts. These target texts can be in the same language as the one of the source corpus, and could present less difficulties in the matching system, or multilingual (Terryn et al., 2018) , in these cases using translations from existing semantic resources could represent a solution.",
"cite_spans": [
{
"start": 213,
"end": 230,
"text": "(Broughton, 2008)",
"ref_id": "BIBREF11"
},
{
"start": 891,
"end": 907,
"text": "(Barri\u00e8re, 2006)",
"ref_id": "BIBREF5"
},
{
"start": 992,
"end": 1010,
"text": "(Condamines, 2018)",
"ref_id": "BIBREF15"
},
{
"start": 1258,
"end": 1288,
"text": "(Loginova Clouet et al., 2012)",
"ref_id": "BIBREF35"
},
{
"start": 1381,
"end": 1405,
"text": "(Daille and Hazem, 2014)",
"ref_id": "BIBREF17"
},
{
"start": 1508,
"end": 1529,
"text": "(Weller et al., 2011)",
"ref_id": "BIBREF47"
},
{
"start": 1655,
"end": 1675,
"text": "(Cabr\u00e9 et al., 2001)",
"ref_id": "BIBREF12"
},
{
"start": 1743,
"end": 1761,
"text": "(Condamines, 2018)",
"ref_id": "BIBREF15"
},
{
"start": 2236,
"end": 2254,
"text": "(Condamines, 2007)",
"ref_id": "BIBREF13"
},
{
"start": 2577,
"end": 2601,
"text": "(Rennesson et al., 2020)",
"ref_id": "BIBREF41"
},
{
"start": 3146,
"end": 3162,
"text": "(Barri\u00e8re, 2006)",
"ref_id": "BIBREF5"
},
{
"start": 3534,
"end": 3555,
"text": "(Terryn et al., 2018)",
"ref_id": "BIBREF45"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Terms Extraction",
"sec_num": "3.1."
},
{
"text": "In this paper, the gold standards taken into account are in Italian language or have been translated in Italian -Nist and Iso -this reflects the native purpose of the project that was intended to provide a guidance for the understanding of the Cybersecurity domain in Italian language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terms Extraction",
"sec_num": "3.1."
},
{
"text": "This paper is going to give a description of the exploited methodologies in automatizing the way thesauri, specifically for the case of study, i.e., Cybersecurity, can be constructed by means of semantic similarity procedures and patterns configuration related to the causative connections. The automatized methodologies used for the configuration of thesauri's structure (Yang and Powers, 2008b; Morin and Jacquemin, 1999) , can quicken the process related to the arrangement of textual relations network to shape the informative tissue of a domain. To achieve this framework system different approaches can be pursued, starting from lexico-syntactic patterns conformation (Condamines, 2007) , and experimenting other solutions such as the ones proposed by (Grefenstette, 1994) with \"Sextant\", or (Kageura et al., 2000) with their methodology in considering the common entries in two different thesauri and constructing pairs of codes. As linguistic structures that are very frequent within a corpus of documents (Lefeuvre, 2017) , patterns allow to discover among terms which are the conceptual relations (Bernier-Colborne and Barri\u00e8re, 2018) . The study of patterns dates way back, at the end of 90' the works of Hearst (1992) were, for instance, firstly focused on the configuration of Noun Phrases followed by other morpho-syntactic structures to be found in texts. Many authors in the literature studied the ways nominal and verbal phrases allow to identify semantic relations between terms through syntagmatic or phrasal structures (Girju et al., 2006) . The typologies of lexico-syntactic markers help in retrieving the desired semantic information about the terminology proper to a specialized domain (Nguyen et al., 2017) , that's the case of the casual relationships between terms. This particular kind of connection is notably described in the works of Barri\u00e8re (2002) in which the author gives a wide-ranging perspective for investigating the causal relationships in informative texts. As the author underlines, it is not an easy task to group the causative verbs that should isolate the representative terms of a domain to be linked through a cause-effect relation. Nevertheless, grouping some of them can help in identifying the semantic associations to be reflected in a controlled vocabulary given the domain-oriented nature of the casual connections. Indeed, retrieving this type of patterns is a context-dependent procedure: in considering the source area of study and having some technical knowledge about it, terminologists can much easily analyse in an autonomous and accurate way a combination of semantic relationships (Condamines, 2008) .",
"cite_spans": [
{
"start": 372,
"end": 396,
"text": "(Yang and Powers, 2008b;",
"ref_id": "BIBREF50"
},
{
"start": 397,
"end": 423,
"text": "Morin and Jacquemin, 1999)",
"ref_id": "BIBREF38"
},
{
"start": 674,
"end": 692,
"text": "(Condamines, 2007)",
"ref_id": "BIBREF13"
},
{
"start": 758,
"end": 778,
"text": "(Grefenstette, 1994)",
"ref_id": "BIBREF21"
},
{
"start": 798,
"end": 820,
"text": "(Kageura et al., 2000)",
"ref_id": "BIBREF30"
},
{
"start": 1014,
"end": 1030,
"text": "(Lefeuvre, 2017)",
"ref_id": "BIBREF34"
},
{
"start": 1129,
"end": 1144,
"text": "Barri\u00e8re, 2018)",
"ref_id": "BIBREF7"
},
{
"start": 1216,
"end": 1229,
"text": "Hearst (1992)",
"ref_id": "BIBREF26"
},
{
"start": 1539,
"end": 1559,
"text": "(Girju et al., 2006)",
"ref_id": "BIBREF20"
},
{
"start": 1710,
"end": 1731,
"text": "(Nguyen et al., 2017)",
"ref_id": "BIBREF39"
},
{
"start": 1865,
"end": 1880,
"text": "Barri\u00e8re (2002)",
"ref_id": null
},
{
"start": 2643,
"end": 2661,
"text": "(Condamines, 2008)",
"ref_id": "BIBREF14"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Relations",
"sec_num": "3.2."
},
{
"text": "For what concerns semantic similarity methods in the literature, they have firstly been applied to single word terms (SWTs) using a variety of approaches such as: lexiconbased approaches (Blondel and Senellart, 2002) , multilingual approaches (Wu and Zhou, 2003; van der Plas and Tiedemann, 2006; Andrade et al., 2013) , distributional approaches (Hagiwara, 2008; Hazem and Daille, 2014) and distributed approaches such in (Mikolov et al., 2013; Bojanowski et al., 2016) . This procedure helps in configuring the associations between terms with respect to synonyms connections retrieved from corpora. On this point, it is important to highlight the relevance of extracting reliable lists of candidate terms that could represent the starting point from which to set up a conceptual modeling of a thesaurus as well as a basis to analyse and define the internal domainspecific synonyms and hyperonyms (Meyer and Mackintosh, 1996) .",
"cite_spans": [
{
"start": 187,
"end": 216,
"text": "(Blondel and Senellart, 2002)",
"ref_id": "BIBREF8"
},
{
"start": 243,
"end": 262,
"text": "(Wu and Zhou, 2003;",
"ref_id": "BIBREF48"
},
{
"start": 263,
"end": 296,
"text": "van der Plas and Tiedemann, 2006;",
"ref_id": "BIBREF46"
},
{
"start": 297,
"end": 318,
"text": "Andrade et al., 2013)",
"ref_id": "BIBREF0"
},
{
"start": 347,
"end": 363,
"text": "(Hagiwara, 2008;",
"ref_id": "BIBREF23"
},
{
"start": 364,
"end": 387,
"text": "Hazem and Daille, 2014)",
"ref_id": "BIBREF24"
},
{
"start": 423,
"end": 445,
"text": "(Mikolov et al., 2013;",
"ref_id": "BIBREF37"
},
{
"start": 446,
"end": 470,
"text": "Bojanowski et al., 2016)",
"ref_id": "BIBREF9"
},
{
"start": 898,
"end": 926,
"text": "(Meyer and Mackintosh, 1996)",
"ref_id": "BIBREF36"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Relations",
"sec_num": "3.2."
},
{
"text": "At this stage, the Italian Cybersecurity thesaurus, on which our paper focuses to describe automatic thesauri construction methodologies, contains 246 terms in the source language (it) and most of them have their definition, or Scope Notes (SN) according to standardized tags (ISO/TC 46/SC 9 2011), taken from the texts from which they derive inside the corpus or the translated gold standards definitions, i.e. Nist and Iso. The thesaurus has been built on the basis of the thesauri construction guidelines from ISO/TC 46/SC 9 2011 and 2013: terms have been formalized in order to guarantee the sharing of information in a standardized way, the concepts of the source corpus have been represented by preferred terms organized according to a network of hierarchical, synonymous and associative semantic relationships. This system allows to set up a knowledge organization oriented towards a creation of semantic connections that, in turn, can create a reflection of the informative scope inside the corpus texts.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus Structure on Cybersecurity",
"sec_num": "4."
},
{
"text": "The structure phase of the thesaurus for Cybersecurity has started by evaluating the list of terms extracted by using a semi-automatic semantic tool, TextToKnowledge (T2K) (Dell'Orletta et al., 2014) , specifically taking into account the frequency scores of the most representative terms and isolating them as being the main candidate terms to be sent to experts' validation process. It was thanks to the co-working process with domain experts that the first list of candidate terms has been filtered out and the first categories, from which the thesaurus structure was developed, provided. This phases resulted after having taken into account several terminological passages:",
"cite_spans": [
{
"start": 172,
"end": 199,
"text": "(Dell'Orletta et al., 2014)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus Structure on Cybersecurity",
"sec_num": "4."
},
{
"text": "\u2022 the matching process between the output lists derived from the semantic extraction and the taxonomies contained in the gold standards of Nist and Iso; these lists of terms have been translated into Italian language by using an automatic translation software, TRA-DOS, and a multi/crosslingual terminological platform, IATE;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus Structure on Cybersecurity",
"sec_num": "4."
},
{
"text": "\u2022 the inverse frequency ranks in the term lists;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus Structure on Cybersecurity",
"sec_num": "4."
},
{
"text": "\u2022 the head-term grouping system T2K processed.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus Structure on Cybersecurity",
"sec_num": "4."
},
{
"text": "In this way, merging the output of a semantic extractor tool, the terminology competencies and the group of experts' validation and supervision, the four main top entry categories have been selected: Cybersecurity, Cyberbullism, Cyber defence, Cybercriminality. The goal of the research activity presented in this paper is to improve the decision-making process towards the thesaurus construction by means of approaches that rely on patterns configuration and semantic similarity measures in order to enrich the informative tissue inside the controlled vocabulary.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thesaurus Structure on Cybersecurity",
"sec_num": "4."
},
{
"text": "In this section the sets of documents from which the candidate terms have been extracted by using several approaches are presented. The first one refers to the Italian gold standard corpus, i.e., Clusit, and the other, i.e., Cybersecurity corpus, is the one used to build on the source corpus. Taking in consideration a highly specialized field of knowledge with plenty of words in English meant to create a shared base of information among users, the terms extrated resulted to be a hybrid syllabous of English and Italian terms. This because the domain of Cybersecurity owns several technical terms that can be maintained in their English version even providing variants, e.g., hackers or exploit.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Corpora",
"sec_num": "5.1."
},
{
"text": "Clusit corpus indicates the reports that have been published by an Italian Cybersecurity organization which shares some of the main cyber threats and attacks together with descriptions, reviews, and a final glossary.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Clusit Corpus",
"sec_num": "5.1.1."
},
{
"text": "Designing a corpus (Leech, 1991) , from which to develop a strong terminological knowledge base that guarantees a rich-context dependency to transmit a reliable representation of a domain, leads to generate a semantic fundamental dataframe that can be representative of the area of study to be analysed (Condamines, 2018) . The Cybersecurity corpus is composed of 220 laws documents and 342 5-sectororiented magazines. The collection of the texts that compose the source corpus is heterogeneous, this means that the information included takes its ground from legislative documents, regulations, norms, directives, guidelines as well as domain-oriented magazines in order to provide an exhaustive resource to assemble the information representation about the field of knowledge. The information included within the divulgative corpus, with respect to the law data set, provided higher accurate terminology, more targeted kind of concepts to be represented with terms. ",
"cite_spans": [
{
"start": 19,
"end": 32,
"text": "(Leech, 1991)",
"ref_id": null
},
{
"start": 303,
"end": 321,
"text": "(Condamines, 2018)",
"ref_id": "BIBREF15"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Cybersecurity Corpus",
"sec_num": "5.1.2."
},
{
"text": "For evaluation, we used five terminological lists:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Clusit The Clusit term list contains the main domain specific terms of the reports gathered in a glossary which is composed by a syllabous of these latter followed by their definitions;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Glossary The Glossary term list contains terms with their definitions published by a political intelligence organism, this characteristic has to be taken into account in considering the accuracy and appropriateness of its derived terminology that seems to be weaker than the other more technical domain-oriented resources;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Nist The Nist 7298 -Glossary of Key Information Security Terms (Kisserl, 2013) term list is a complex of terms alphabetically ordered and accompanied by their definitions, also derived from other reference standards. It's considered as a main authoritative data set for Cybersecurity experts on the same level as the Iso list;",
"cite_spans": [
{
"start": 63,
"end": 78,
"text": "(Kisserl, 2013)",
"ref_id": "BIBREF31"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Iso The Iso term list refers to the International Standard (ISO/IEC 27000, 2016) for Security and Technology, and it contains, as the Nist, the terms alphabetically ordered with their definitions;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Cyber The Cybersecurity term list contains candidate terms taken from the post-processed texts connected together through the main semantic relationships proper to thesauri (Broughton, 2008) , i.e., hierarchical, synonymy, association. These relations are respectively formalized by standard tags (ISO/TC 46/SC 9 2011 and 2013):",
"cite_spans": [
{
"start": 173,
"end": 190,
"text": "(Broughton, 2008)",
"ref_id": "BIBREF11"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "broader term broader term (BT) that stands for hyperonyms;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "narrower term narrower term (NT) that stands for hyponyms;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "used for used for (UF) and use (USE) that represent the synonymy relation;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "related term related term (RT).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Hereafter some examples of the four addressed relations: hyperonymy (Hyp), synonymy (Syn), related terms (Rel) and cause (Cause).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Hypernym Spam/Phishing, Spam/Smishing, Crypto miner malware/Bitcoin Virus, DoS/DDoS;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Synonym Crackers/Black hat, Sotware malevoli (malicious software)/Malware, Cyber minacce (cyber threats)/Cyber Threat Actors;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Related Blockchain/Propriet\u00e0 di sicurezza (security properties), Crackers/Hacking, Cyber defence/Cybersecurity;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Causative verb Spoofing/Attacchi informatici (cyber attacks) (to alterate), Integrit\u00e0 (integrity)/Cyber minacce (cyber threats) (to damage), Attacco (attack)/Malware (implicate).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Tables 2 and 3 respectively illustrate the size of the term evaluation lists and the distributions of each semantic relation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "Clusit Glossary Nist Iso Cyber #terms 202 284 1282 89 247 ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Terminology Lists",
"sec_num": "5.2."
},
{
"text": "In this section we provide a description of the chosen tools to execute the terminology extraction.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Term Extraction Tools",
"sec_num": "6.1."
},
{
"text": "TermSuite (Cram and Daille, 2016 ) is a toolkit for terminology extraction and multilingual term alignment. Its performance is quite immediate when it runs over big data sets. The term extraction provided by TermSuite is a list of representative terms that are presented together with different properties, e.g., their frequency, accuracy, specificity. Terms are therefore ordered according to their unithood and application to the domain. One of the main feature that shapes the quality of this software is its syntactic and morphological variants detection among terms, e.g, lexical reduction, composition, coordination, derivation (Lanza and Daille, 2019) . Variants identification given by the output list in TermSuite represents one of the methods selected to retrieve hyperonyms as well as synonyms in the source corpus. In fact, through the denominative, conceptual and linguistic variants included in the terminological output it is possible to detect in which ways terms are expanded by other semantic elements, reduced, related to an opposite one, or appearing in several linguistic conformations, e.g., cyber security or cybersecurity. Below a list of few examples to show the variations given by the outputs in TermSuite terminological extraction for Cybersecurity domain in Italian language that can help in detecting semantic associations to be included in the thesaurus:",
"cite_spans": [
{
"start": 10,
"end": 32,
"text": "(Cram and Daille, 2016",
"ref_id": "BIBREF16"
},
{
"start": 634,
"end": 658,
"text": "(Lanza and Daille, 2019)",
"ref_id": "BIBREF32"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "TermSuite -Variants Detection Tool",
"sec_num": "6.1.1."
},
{
"text": "\u2022 denominative variants: NPN: hacker (21 matches) del telefono (mobile hacker)\u2192 NA: hacker telefonico",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "TermSuite -Variants Detection Tool",
"sec_num": "6.1.1."
},
{
"text": "\u2022 conceptual variants: NPN: worm (8 matches) \u2192 NPNPNA: worm del genere del famigerato nimda (worm, the infamous nimda kind one)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "TermSuite -Variants Detection Tool",
"sec_num": "6.1.1."
},
{
"text": "\u2022 linguistic variants: N: antivirus (6 matches) \u2192 A: anti-virus",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "TermSuite -Variants Detection Tool",
"sec_num": "6.1.1."
},
{
"text": "In the next paragraphs we show how these terms included in the examples above are returned in different ways by the other systems, T2K and PKE.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "TermSuite -Variants Detection Tool",
"sec_num": "6.1.1."
},
{
"text": "T2K is an Italian software to automatically extract linguistic information from domain-oriented data sets (Dell'Orletta et al., 2014) . The software takes a corpus and processes it according to a default or customized configuration given in input. The list of terms is sorted by the inverse frequency measure or indexed by grouping them according to head-terms ordering. One of the advantages of this semantic extractor is the possibility to personalize the patterns meant to be exploited to execute the extraction of domain-oriented terminology; in this way a more precise semantic chains output can be achieved. On the other hand, though this software shows many benefits related to its flexibility in adapting the configuration to the terminology needs, it performs very slowly when it comes to analyse big corpora. Also for T2K we provide a small set of terms that appear differently from TermSuite's output, or are given with a larger number of results (this is because in T2K the terminological extraction is numerically higher than TermSuite) referred to the aforementioned examples. They provide as well some extra information that can help in orientating the structure outline of the thesaurus blocks:",
"cite_spans": [
{
"start": 106,
"end": 133,
"text": "(Dell'Orletta et al., 2014)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "\u2022 hacker (519 matches) \u2192 hackeraggio (hacking)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "\u2022 worm (102 matches) \u2192 worm via posta elettronico (worm via e-mail)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "\u2022 antivirus (127 matches) \u2192 antivirus affetto da trojan (antivirus affected by trojan) 6.1.3. Pke -Keyphrases Identification Tool PKE (Boudin, 2016) is an open-source python keyphrase extraction toolkit that implements several keyphrase extraction approaches. From a linguistic point of view, PKE resulted to be very efficient in terms of providing a semiautomatic structuring of information since many candidate terms, which have been selected as being part of the Cybersecurity thesaurus, are grouped alongside with other ones that, in turn, could represent their associative semantic chains. For this section we provide as well related examples for the terms outputs precision:",
"cite_spans": [
{
"start": 134,
"end": 148,
"text": "(Boudin, 2016)",
"ref_id": "BIBREF10"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "\u2022 hacker and worm are found in a same keyphrase cluster \u2192 sistemi (systems), rete (network), worm analisi (worm analysis), password, hacker",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "\u2022 antivirus/anti-virus not present New information is on the other hand given by terms that are not appearing in the previous two extractors and that are grouped in a way that can help in structuring their relations inside the thesaurus' outline. In the following cluster it can be observed how the candidate complex term cyber counterintelligence could be organized according to the surrounding terms that help in conceiving it as a technique or a procedure in the cyber intelligence and cyber defence tasks. 6.1.4. BERT Feature-based approaches are often used for automatic term extraction (Terryn et al., 2018) . However, it is often time consuming and not always straightforward to design the most appropriate features to efficiently train a classifier. In order to get rid of the handcrafted features, we chose to apply, as an alternative, a very recent deep neural network approach: Bidirectional Encoder Representations from Transformers (BERT). BERT has proven to be efficient in many downstream NLP tasks (Devlin et al., 2018) including next sentence prediction, question answering, name entity recognition (NER), etc. BERT can be used for feature extraction or for classification. In automatic term extraction (ATE) task, we use BERT as a binary classifier for term prediction. The main idea is to associate each term with its context. Hence, by analogy to next sentence prediction, the first sentence given to BERT is the one which contains the term, and the sentence to predict is the term itself. For training, we feed the model with all the context/term pairs that appear in the corpus as positive examples. The negative examples are generated randomly. Therefore, we hypothesize Table 5 : Terminology extraction results of T2K, TermSuite, PKE and BERT on the Clusit and Cybersecurity corpora. The evaluation is conducted on five lists (Clusit, Glossary, Nist, Iso and Cyber) and the results (%) are given in terms of Precision (P), Recall (R) and F-measure (F1).",
"cite_spans": [
{
"start": 592,
"end": 613,
"text": "(Terryn et al., 2018)",
"ref_id": "BIBREF45"
},
{
"start": 1014,
"end": 1035,
"text": "(Devlin et al., 2018)",
"ref_id": "BIBREF19"
}
],
"ref_spans": [
{
"start": 1694,
"end": 1701,
"text": "Table 5",
"ref_id": null
}
],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "that BERT can learn associations between terms and their contexts. Term extraction systems, with the exception of BERT, include filtering methods that allow the user to set thresholds on various statistical measures above which the ranked candidate terms are kept. In order to favour recall, we decided not to apply any further filtering except for those included as default parameters. Table 4 shows the number of extracted candidates for each used tool/method. T2K software outputs come out with the largest terminological range sets and BERT with the smallest.",
"cite_spans": [],
"ref_spans": [
{
"start": 387,
"end": 394,
"text": "Table 4",
"ref_id": "TABREF5"
}
],
"eq_spans": [],
"section": "T2K -Language Design Tool",
"sec_num": "6.1.2."
},
{
"text": "We conduct an evaluation on five terminological lists: Clusit, Glossary, N ist, Iso and Cyber and on two corpora: Clusit (Clusit) and Cybersecurity (Cyber). The results are given in terms of Precision (P), Recall (R) and F-measure (F1). We also give the coverage of each list on each corpus. Table 5 illustrates the obtained results on the terminology extraction task. Overall, we observe weak results for all the methods. Nonetheless, the recall is much higher especially for PKE and T2K which correlates with the number of their output candidates (see Table 4 ). The evaluation's list size is very small (around 200) and systems output is often around thousands of terms, which explains the very low precision. Moreover, the evaluation lists are not exhaustive and, by consequence, do not allow a fair evaluation on precision. Indeed, several correct terms which are not present in the evaluation lists have been observed. Finally, based on the F1 score, BERT obtained the best results in all the cases.",
"cite_spans": [],
"ref_spans": [
{
"start": 292,
"end": 299,
"text": "Table 5",
"ref_id": null
},
{
"start": 554,
"end": 561,
"text": "Table 4",
"ref_id": "TABREF5"
}
],
"eq_spans": [],
"section": "Term Extraction Experiments",
"sec_num": "6.2."
},
{
"text": "To address the semantic similarity task, we introduce in the following sections pattern-based and word embeddingbased approaches.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Semantic Relations Automatization",
"sec_num": "7."
},
{
"text": "Among the approaches which have been used for the development of this strategy that could retrieve the semantic connections starting from a domain-oriented data set, the patterns recognition has been one of them (R\u00f6siger et al., 2016) . For the purposes of this research activity, some key verbs have been taken into account to represent the causative relationships among the terms included in all the documents of the Italian Cybersecurity source corpus. Almost all of these first verbs imply a relation of agent -cause that provokes some circumstances. The objective of this path-based configuration is to improve the accuracy of the associative relationships included in thesauri and labelled as RT, which stands for Related Terms (ISO/TC 46/SC 9, 2011). Indeed, as stated in (R\u00f6siger et al., 2016) work on the achievement of good sets of semantic relationships by employing NLP techniques, the decision of certain verbobject pairs relies on the domain pertinence and relevance, and also on the assumption that these pairs can be syntactically correct. In this step, the verbs considered to launch the queries meant to group the causative relationships among the candidate terms has not followed frequency drills. Almost thirty of the most common casual verbs in Italian have been exploited to retrieve the co-occurrences in the source corpus. The aim about using patterns configuration related to the causative relations (Lefeuvre and Condamines, 2015) is that of providing an improvement in the structure of the related terms in the thesaurus. In ISO Standard 25964 of 2013, when it comes to discuss about the interoperability of the systems, the associative mapping is described as a connection that \"[...] may be established between concepts when they do not qualify for equivalence or hierarchical mappings, but are semantically associated to such an extent that documents indexed with the one are likely to be relevant in a search for the other.\" As can be further observed, the associative relationship in thesauri systematization is among the others, hierarchical and equivalence, the one that presents more ambiguity in the way it connects the domain-oriented terms. By using causative-based patterns the references from one specific term to another seem more precise and reliable. The following list presents some examples for the selected causative verbs, some of these relations added new information about the connections among the Cybersecurity specicialized terms, i.e., the relation that occurs between camouflage and password, or cyber threats and the security properties; sometimes they confirmed the already configured outline of the thesaurus, as cyber attacks and DDoS or spoofing.",
"cite_spans": [
{
"start": 212,
"end": 234,
"text": "(R\u00f6siger et al., 2016)",
"ref_id": "BIBREF43"
},
{
"start": 779,
"end": 801,
"text": "(R\u00f6siger et al., 2016)",
"ref_id": "BIBREF43"
},
{
"start": 1425,
"end": 1456,
"text": "(Lefeuvre and Condamines, 2015)",
"ref_id": "BIBREF33"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "\u2022 provocare (to provoke): virus -worm cyber attacks -DDoS risks -cyber threats \u2022 danneggiare (to damage):",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "crackers -data cyber threats -integrity, privacy, availability \u2022 comportare (to imply):",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "cyber attacks -malware cyber harrassment -cyber bullism \u2022 alterare (to alter):",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "camouflage -password spoofing -cyber attacks \u2022 manomesso da (sabotaged by):",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "monitoring -cyber attacks monitoring -DoS \u2022 impattare (to impact):",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "DDoS -cyber attacks monitoring -cybersecurity",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "In summary, causative connections retrieved from source corpus provided added information to the existing ones contained in the Italian Cybersecurity thesaurus, which have already gone through an evaluation phase by a group of experts of the domain.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Patterns-based",
"sec_num": "7.1."
},
{
"text": "Word embedding models have been showing to be very effective in word representation. They have been applied in several NLP tasks including word disambiguation, semantic similarity, bilingual lexicon induction (Mikolov et al., 2013; Arora et al., 2017; Bojanowski et al., 2016) , etc. For semantic similarity, and more precisely synonym extraction of multi-word terms, two compositionality-based techniques have been proposed (Hazem and Daille, 2018) . The first technique called Semi-compositional word embeddings is based on distributional analysis (Hazem and Daille, 2014) and assumes that the head or a tail is shared by two semantically related terms. The second technique called Full-compositional word embeddings is inspired by the idea that phrases can be represented by an element-wise sum of the word embeddings of semantically related words of its parts (Arora et al., 2017) . In our experiments we follow the principle of the second technique and apply it to the automatic extraction of hyperonyms, synonyms, related and causative terms. The idea is to answer the question: are word embedding models able to extract semantic relations using full-compositionality? All the multi-word terms (MWTs) are represented by a single embedding vector. Each MWT is first characterized by an element-wise sum of its word embedding elements. Then, the cosine similarity measure is applied to extract MWTs synonyms, hypernyms, causative and related terms.",
"cite_spans": [
{
"start": 209,
"end": 231,
"text": "(Mikolov et al., 2013;",
"ref_id": "BIBREF37"
},
{
"start": 232,
"end": 251,
"text": "Arora et al., 2017;",
"ref_id": "BIBREF1"
},
{
"start": 252,
"end": 276,
"text": "Bojanowski et al., 2016)",
"ref_id": "BIBREF9"
},
{
"start": 425,
"end": 449,
"text": "(Hazem and Daille, 2018)",
"ref_id": "BIBREF25"
},
{
"start": 550,
"end": 574,
"text": "(Hazem and Daille, 2014)",
"ref_id": "BIBREF24"
},
{
"start": 864,
"end": 884,
"text": "(Arora et al., 2017)",
"ref_id": "BIBREF1"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Word Embedding-based",
"sec_num": "7.2."
},
{
"text": "We evaluate two word embedding models: word2vec (W2V) (Mikolov et al., 2013) and fastText (Bojanowski et al., 2016) . For both models we experiment the Skipgram (Sg) and the Continuous Bag of Words (CBOW) models. The results are shown in terms of precision at 100 (P@100). Table 6 : Results of semantic relation extraction of word2vec (W2V) and fastText using the Precision at 100 (P@100%) score.",
"cite_spans": [
{
"start": 54,
"end": 76,
"text": "(Mikolov et al., 2013)",
"ref_id": "BIBREF37"
},
{
"start": 90,
"end": 115,
"text": "(Bojanowski et al., 2016)",
"ref_id": "BIBREF9"
}
],
"ref_spans": [
{
"start": 273,
"end": 280,
"text": "Table 6",
"ref_id": null
}
],
"eq_spans": [],
"section": "Semantic Similarity Experiments",
"sec_num": "7.3."
},
{
"text": "As illustrated in Table 6 , all the models fail to extract hypernyms, related, and causative relations. Only synonym extraction exibits acceptable results with Sg (45.7%). Nonetheless, the weak results, even for synonyms can be explained by the mixed nature of language in the Cybersecurity corpus terminology. Indeed, several terms are in English and their related terms in Italian or conversely. This circumstance might weaken the embedding models for low frequency terms.",
"cite_spans": [],
"ref_spans": [
{
"start": 18,
"end": 25,
"text": "Table 6",
"ref_id": null
}
],
"eq_spans": [],
"section": "Semantic Similarity Experiments",
"sec_num": "7.3."
},
{
"text": "To draw guidelines for automatic thesaurus construction, we discuss the following questions: (i) which term extraction system to use; (ii) which system output is the most convenient to enrich an existing term list; (iii) which word embedding model is the most suitable for semantic relation extraction; and, finally, (iv) what kind of relations are extracted by word embedding models. As stated in previous work (Terryn et al., 2018) , the evaluation of automatic term extraction is not an easy task. This observation is confirmed in this paper with regards to the obtained results on different evaluation lists (Clusit, Cyber, Iso, etc.) . This is particularly true because our evaluation lists are not exhaustive and, for this reason, they don't reflect a real term extraction evaluation scenario. However, they do reflect the situation of thesaurus enrichment, which we stress in this work. If we cannot draw final conclusions on the term extraction performance of the evaluated systems, we can still observe their weak performance on the addressed small subset of terms on Cybersecurity. Nonetheless, this result is to be counterbalanced by encouraging new terms extracted by these systems. Indeed, a manual evaluation of BERT system output, for instance, has shown many new accurate extracted terms. This work represents the first attempt to use BERT model for terminology extraction. Overall, BERT obtained the best results with minimum supervision and no pattern analysis. This is encouraging since no careful filtering process has been applied, and opens the path for new strategies to pursue for term extraction using deep neural approaches. For what concerns the types of relations extracted by word embedding models, for the most part the terms in the lists referred to the three semantic relations categories, i.e., hierarchy, association and causative links, prove to be quite similar in the occurrences they provided and, at times, not very faithful, e.g., cyber gang is connected in an hierarchical way with criptography. On the other hand, the synonyms detection showed better results and the findings are very exhaustive both for what concerns the retrieval of the synonyms themselves, and for the recognition, among the outputs given by the models, of other candidate related terms to add in the thesaurus. The connections given by these models were performed using the existing thesaurus relations, which have been created following the ISO 25964:2011 rules, as source correspondences to be enhanced with sophisticated grouping procedures. Though the manual evaluation of these series of interrelations has inferred quite similar proximity among the terms extracted in all the four classes of relations, at least on a quantitative level, e.g. rischi cyber (cyber risks), anti spam, hackeraggio (hacking) appear for almost all the cases, many associated terms helped in improving the thesaural systematization. It should be underlined that when evaluating these kind of lists, a minimum level of knowledge expertise about the technical domain to be studied is required since many terms connected with the head ones sometimes appear related in a very implied way, at least for the domain experts, e.g., cavalli di troia (trojan horse) or zero-day. We provide few examples of the additional inputs provided by word embedding techniques on the Italian source corpus about Cybersecurity. It is implied that a new evaluation from the experts of the domain is necessary for the seek of reaching out high pertinence and accuracy levels in the terminological enhanced network meant to transposed in the semantic tool, which is supposed to be shared.",
"cite_spans": [
{
"start": 412,
"end": 433,
"text": "(Terryn et al., 2018)",
"ref_id": "BIBREF45"
},
{
"start": 612,
"end": 638,
"text": "(Clusit, Cyber, Iso, etc.)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "Hyperonyms detection 1. gestione del rischio cyber (risk management) which has as hyponym piano di risposta al rischio cyber (risk response measures), has been connected with: attacchi cibernetici (cyber attacks), cavalli di troia (trojan horse), cyber intelligence, difesa informatica (cyber security); this confirms the thesaurus outline regarding the top term category of cybersecurity and adds another one to be considered, i.e., cyber intelligence.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "2. intrusion detection system -host-based, in the thesaurus is the hyponym of network security systems.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "Among the terms related in a hierarchical way, network security systems has been confirmed, and, in turn, other related terms have been included in the semantic structure, e.g., hacker, mid hacking, sniffing and malware.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "Synonymys detection 1. cybersecurity has been related to the following synonyms that can be considered as positive candidates for the thesaurus: difesa informatica (informative defence), deep security, sicurezza cibernetica (cibernetic security), protezione cibernetica (cibernetic protection), sicurezza dei sistemi informativi (informative systems security), sicurezza ict (ict security).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "2. malware has been found related with these synonyms: software malevolo (malicious software), programmi malevoli (malicious programs), confirming the existing synonymous structure in the thesaurus; the interesting result is that malware is associated in the same list with several representative terms that will be, in a future perspective, conceived as candidates to improve its semantic connections: spyware, keylogger, firewall, exploit.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "Related terms detection 1. zero-day that in the thesaurus is connected on an associative level with software vulnerabilities, is grouped together with trojan horse, anti spam, hacking, privacy, risk management.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "2. cyber molestie (cyber harassment), related in the thesaurus, among others, with cyber stalking, has an improved structuring matches since it is found associated also with cyber theft, hacking, threats, cyber insurance.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "Causative relations detection 1. cyber minacce (cyber threats) was connected through the causative verb to damage to the security properties of data, in these models it is linked to cyber intelligence, difesa informatica (cyber security), hackeraggio (hacking) and cavalli di troia (trojan horse).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "2. bitcoin was associated with data loss through the causative pattern verb to prevent, with the application of these embedding techniques it seems also related with risk management, cyber risk, spam, hacking.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discussion",
"sec_num": "8."
},
{
"text": "Automatic thesaurus construction requires efficient methods to collect terminologies and to structure them in a representative way. We discussed in the present paper different approaches for the two building blocks of thesaurus construction: (i) term extraction and (ii) similarity linking. We conducted experiments on an Italian Cybersecurity corpus and reported the performance of existing methods with regards to several evaluation lists. We also proposed a new BERT-based approach that outperformed existing methods on the task of term extraction. If on a general perspective the obtained results provided not so high scores, we observed that system outputs contain accurate candidates that can be used to enrich the existing thesaurus. This is noticeable for the proposed BERT model. Also, regarding semantic similarity, word embedding models showed interesting outputs especially for synonyms and causative relations.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "9."
}
],
"back_matter": [],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Synonym acquisition using bilingual comparable corpora",
"authors": [
{
"first": "D",
"middle": [],
"last": "Andrade",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Tsuchida",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "Onishi",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Ishikawa",
"suffix": ""
}
],
"year": 2013,
"venue": "International Joint Conference on Natural Language Processing (IJCNLP'13)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Andrade, D., Tsuchida, M., Onishi, T., and Ishikawa, K. (2013). Synonym acquisition using bilingual compara- ble corpora. In International Joint Conference on Natu- ral Language Processing (IJCNLP'13), Nagoya, Japan.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "A simple but tough to beat baseline for sentence embeddings",
"authors": [
{
"first": "S",
"middle": [],
"last": "Arora",
"suffix": ""
},
{
"first": "L",
"middle": [],
"last": "Yingyu",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Tengyu",
"suffix": ""
}
],
"year": 2017,
"venue": "Proceedings of the 17th International Conference on Learning Representations (ICLR'17)",
"volume": "",
"issue": "",
"pages": "1--11",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Arora, S., Yingyu, L., and Tengyu, M. (2017). A sim- ple but tough to beat baseline for sentence embeddings. In Proceedings of the 17th International Conference on Learning Representations (ICLR'17), pages 1-11.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Modeling resources and capabilities in enterprise architecture: A well-founded ontology-based proposal for archimate. Information systems",
"authors": [
{
"first": "C",
"middle": [],
"last": "Azevedo",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Iacob",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Almeida",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Van Sinderen",
"suffix": ""
},
{
"first": "L",
"middle": [],
"last": "Ferreira Pires",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Guizzardi",
"suffix": ""
}
],
"year": 2015,
"venue": "",
"volume": "54",
"issue": "",
"pages": "235--262",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Azevedo, C., Iacob, M., Almeida, J., van Sinderen, M., Ferreira Pires, L., and Guizzardi, G. (2015). Model- ing resources and capabilities in enterprise architecture: A well-founded ontology-based proposal for archimate. Information systems, 54:235-262, 12.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Hierarchical refinement and representation of the causal relation",
"authors": [
{
"first": "C",
"middle": [],
"last": "Barri\u00e8re",
"suffix": ""
}
],
"year": 2002,
"venue": "Terminology",
"volume": "8",
"issue": "1",
"pages": "91--111",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Barri\u00e8re, C. (2002a). Hierarchical refinement and repre- sentation of the causal relation. Terminology, 8(1):91- 111.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Investigating the causal relation in informative texts",
"authors": [
{
"first": "C",
"middle": [],
"last": "Barri\u00e8re",
"suffix": ""
}
],
"year": 2002,
"venue": "Terminology",
"volume": "7",
"issue": "4",
"pages": "135--154",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Barri\u00e8re, C. (2002b). Investigating the causal relation in informative texts. Terminology, 7(4):135-154.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Text-Based Studies in honour of Ingrid Meyer, Lexicography, Terminology and Translation, chapter 5",
"authors": [
{
"first": "C",
"middle": [],
"last": "Barri\u00e8re",
"suffix": ""
}
],
"year": 2006,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Barri\u00e8re, C. (2006). Semi-automatic corpus construction from informative texts. In Lynne Bowkes, editor, Text- Based Studies in honour of Ingrid Meyer, Lexicography, Terminology and Translation, chapter 5. University of Ottawa Press, January.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "Methodologies for data quality assessment and improvement",
"authors": [
{
"first": "C",
"middle": [],
"last": "Batini",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Cappiello",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Francalanci",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Maurino",
"suffix": ""
}
],
"year": 2009,
"venue": "ACM Comput. Surv",
"volume": "41",
"issue": "3",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Batini, C., Cappiello, C., Francalanci, C., and Maurino, A. (2009). Methodologies for data quality assessment and improvement. ACM Comput. Surv., 41(3), July.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "CRIM at semeval-2018 task 9: A hybrid approach to hypernym discovery",
"authors": [
{
"first": "G",
"middle": [],
"last": "Bernier-Colborne",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Barri\u00e8re",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of The 12th International Workshop on Semantic Evaluation, SemEval@NAACL-HLT",
"volume": "",
"issue": "",
"pages": "725--731",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Bernier-Colborne, G. and Barri\u00e8re, C. (2018). CRIM at semeval-2018 task 9: A hybrid approach to hypernym discovery. In Proceedings of The 12th International Workshop on Semantic Evaluation, SemEval@NAACL- HLT, New Orleans, Louisiana, June 5-6, 2018, pages 725-731.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Automatic extraction of synonyms in a dictionary",
"authors": [
{
"first": "V",
"middle": [
"D"
],
"last": "Blondel",
"suffix": ""
},
{
"first": "P",
"middle": [],
"last": "Senellart",
"suffix": ""
}
],
"year": 2002,
"venue": "SIAM Workshop on Text Mining",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Blondel, V. D. and Senellart, P. (2002). Automatic extrac- tion of synonyms in a dictionary. In SIAM Workshop on Text Mining.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Enriching word vectors with subword information",
"authors": [
{
"first": "P",
"middle": [],
"last": "Bojanowski",
"suffix": ""
},
{
"first": "E",
"middle": [],
"last": "Grave",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Joulin",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "Mikolov",
"suffix": ""
}
],
"year": 2016,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Bojanowski, P., Grave, E., Joulin, A., and Mikolov, T. (2016). Enriching word vectors with subword informa- tion. CoRR, abs/1607.04606.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "pke: an open source python-based keyphrase extraction toolkit",
"authors": [
{
"first": "F",
"middle": [],
"last": "Boudin",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: System Demonstrations",
"volume": "",
"issue": "",
"pages": "69--73",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Boudin, F. (2016). pke: an open source python-based keyphrase extraction toolkit. In Proceedings of COLING 2016, the 26th International Conference on Computa- tional Linguistics: System Demonstrations, pages 69-73, Osaka, Japan, December. The COLING 2016 Organizing Committee.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "Costruire Thesauri: strumenti per indicizzazione e metadati semantici",
"authors": [
{
"first": "W",
"middle": [],
"last": "Broughton",
"suffix": ""
}
],
"year": 2008,
"venue": "EditriceBibliografica",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Broughton, W. (2008). Costruire Thesauri: strumenti per indicizzazione e metadati semantici\". EditriceBib- liografica, 2008, Milano, Italia Cliffs, NJ.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Automatic term detection: A review of current systems",
"authors": [
{
"first": "M",
"middle": [
"T"
],
"last": "Cabr\u00e9",
"suffix": ""
},
{
"first": "R",
"middle": [
"E"
],
"last": "Bagot",
"suffix": ""
},
{
"first": "J",
"middle": [
"V"
],
"last": "Platresi",
"suffix": ""
}
],
"year": 2001,
"venue": "of Natural Language Processing",
"volume": "",
"issue": "",
"pages": "53--88",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Cabr\u00e9, M. T., Bagot, R. E., and Platresi, J. V. (2001). Au- tomatic term detection: A review of current systems. In Recent Advances in Computational Terminology, vol- ume 2 of Natural Language Processing, pages 53-88. John Benjamins.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "L'interpr\u00e9tation en s\u00e9mantique de corpus : le cas de la construction de terminologies",
"authors": [
{
"first": "A",
"middle": [],
"last": "Condamines",
"suffix": ""
}
],
"year": 2007,
"venue": "",
"volume": "XII",
"issue": "",
"pages": "39--52",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Condamines, A. (2007). L'interpr\u00e9tation en s\u00e9mantique de corpus : le cas de la construction de terminolo- gies. Revue fran\u00e7aise de linguistique appliqu\u00e9e, Vol. XII(2007/1):39-52.",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Taking Genre into account when Analyzing Conceptual Relation Patterns. Corpora",
"authors": [
{
"first": "A",
"middle": [],
"last": "Condamines",
"suffix": ""
}
],
"year": 2008,
"venue": "",
"volume": "8",
"issue": "",
"pages": "115--140",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Condamines, A. (2008). Taking Genre into account when Analyzing Conceptual Relation Patterns. Corpora, 8:115-140.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Terminological knowledge bases from texts to terms, from terms to texts",
"authors": [
{
"first": "A",
"middle": [],
"last": "Condamines",
"suffix": ""
}
],
"year": 2018,
"venue": "The Routledge Handbook of Lexicography",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Condamines, A. (2018). Terminological knowledge bases from texts to terms, from terms to texts. In The Rout- ledge Handbook of Lexicography. Routledge.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Terminology extraction with term variant detection",
"authors": [
{
"first": "D",
"middle": [],
"last": "Cram",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Daille",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of ACL-2016 System Demonstrations",
"volume": "",
"issue": "",
"pages": "13--18",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Cram, D. and Daille, B. (2016). Terminology extraction with term variant detection. In Proceedings of ACL-2016 System Demonstrations, pages 13-18, Berlin, Germany, August. Association for Computational Linguistics.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "Semi-compositional method for synonym extraction of multi-word terms",
"authors": [
{
"first": "B",
"middle": [],
"last": "Daille",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Hazem",
"suffix": ""
}
],
"year": 2014,
"venue": "Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)",
"volume": "",
"issue": "",
"pages": "1202--1207",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Daille, B. and Hazem, A. (2014). Semi-compositional method for synonym extraction of multi-word terms. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14), pages 1202-1207, Reykjavik, Iceland, May. European Lan- guage Resources Association (ELRA).",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "T2K: a system for automatically extracting and organizing knowledge from texts",
"authors": [
{
"first": "F",
"middle": [],
"last": "Dell'orletta",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Venturi",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Cimino",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Montemagni",
"suffix": ""
}
],
"year": 2014,
"venue": "Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Dell'Orletta, F., Venturi, G., Cimino, A., and Montemagni, S. (2014). T2K: a system for automatically extract- ing and organizing knowledge from texts. In Proceed- ings of the Ninth International Conference on Language Resources and Evaluation (LREC'14), Reykjavik, Ice- land, may. European Language Resources Association (ELRA).",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "BERT: pre-training of deep bidirectional transformers for language understanding",
"authors": [
{
"first": "J",
"middle": [],
"last": "Devlin",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Chang",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Lee",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Toutanova",
"suffix": ""
}
],
"year": 2018,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Devlin, J., Chang, M., Lee, K., and Toutanova, K. (2018). BERT: pre-training of deep bidirectional transformers for language understanding. CoRR, abs/1810.04805.",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Automatic discovery of part-whole relations",
"authors": [
{
"first": "R",
"middle": [],
"last": "Girju",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Badulescu",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Moldovan",
"suffix": ""
}
],
"year": 2006,
"venue": "Computational Linguistics",
"volume": "32",
"issue": "1",
"pages": "83--135",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Girju, R., Badulescu, A., and Moldovan, D. (2006). Auto- matic discovery of part-whole relations. Computational Linguistics, 32(1):83-135.",
"links": null
},
"BIBREF21": {
"ref_id": "b21",
"title": "Explorations in Automatic Thesaurus Discovery",
"authors": [
{
"first": "G",
"middle": [],
"last": "Grefenstette",
"suffix": ""
}
],
"year": 1994,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Grefenstette, G. (1994). Explorations in Automatic The- saurus Discovery. Kluwer Academic Publisher, Boston, MA.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Automatic thesaurus construction by machine learning from retrieval sessions",
"authors": [
{
"first": "U",
"middle": [],
"last": "G\u00fcntzer",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "J\u00fcttner",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Seegm\u00fcller",
"suffix": ""
},
{
"first": "F",
"middle": [],
"last": "Sarre",
"suffix": ""
}
],
"year": 1989,
"venue": "Inf. Process. Manage",
"volume": "25",
"issue": "3",
"pages": "265--273",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "G\u00fcntzer, U., J\u00fcttner, G., Seegm\u00fcller, G., and Sarre, F. (1989). Automatic thesaurus construction by machine learning from retrieval sessions. Inf. Process. Manage., 25(3):265-273, May.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "A supervised learning approach to automatic synonym identification based on distributional features",
"authors": [
{
"first": "M",
"middle": [],
"last": "Hagiwara",
"suffix": ""
}
],
"year": 2008,
"venue": "Proceedings of the ACL-08: HLT Student Research Workshop",
"volume": "",
"issue": "",
"pages": "1--6",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hagiwara, M. (2008). A supervised learning approach to automatic synonym identification based on distributional features. In Proceedings of the ACL-08: HLT Student Research Workshop, pages 1-6, Columbus, Ohio, June. Association for Computational Linguistics.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "Semi-compositional method for synonym extraction of multi-word terms",
"authors": [
{
"first": "A",
"middle": [],
"last": "Hazem",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Daille",
"suffix": ""
}
],
"year": 2014,
"venue": "Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hazem, A. and Daille, B. (2014). Semi-compositional method for synonym extraction of multi-word terms. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14), Reyk- javik, Iceland, may. European Language Resources As- sociation (ELRA).",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Word Embedding Approach for Synonym Extraction of Multi-Word Terms",
"authors": [
{
"first": "A",
"middle": [],
"last": "Hazem",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Daille",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hazem, A. and Daille, B. (2018). Word Embedding Ap- proach for Synonym Extraction of Multi-Word Terms. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), Miyazaki, Japan, May 7-12, 2018. European Language Resources Association (ELRA).",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Automatic acquisition of hyponyms from large text corpora",
"authors": [
{
"first": "M",
"middle": [
"A"
],
"last": "Hearst",
"suffix": ""
}
],
"year": 1992,
"venue": "The 15th International Conference on Computational Linguistics",
"volume": "2",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hearst, M. A. (1992). Automatic acquisition of hyponyms from large text corpora. In COLING 1992 Volume 2: The 15th International Conference on Computational Lin- guistics.",
"links": null
},
"BIBREF27": {
"ref_id": "b27",
"title": "Information technology -Security techniques -Information security management systems -Overview and vocabulary. International Standard",
"authors": [],
"year": 2016,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "ISO/IEC 27000, (2016). Information technology -Secu- rity techniques -Information security management sys- tems -Overview and vocabulary. International Stan- dard, February.",
"links": null
},
"BIBREF28": {
"ref_id": "b28",
"title": "ISO 25964-1:2011 Information and documentation -Thesauri and interoperability with other vocabularies -Part 1: Thesauri for information retrieval",
"authors": [],
"year": 2011,
"venue": "International Standard",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "ISO/TC 46/SC 9, (2011). ISO 25964-1:2011 Information and documentation -Thesauri and interoperability with other vocabularies -Part 1: Thesauri for information retrieval. International Standard, August.",
"links": null
},
"BIBREF29": {
"ref_id": "b29",
"title": "ISO 25964-2:2013 Information and documentation -Thesauri and interoperability with other vocabularies -Part 2: Interoperability with other vocabularies. International Standard",
"authors": [],
"year": 2013,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "ISO/TC 46/SC 9, (2013). ISO 25964-2:2013 Information and documentation -Thesauri and interoperability with other vocabularies -Part 2: Interoperability with other vocabularies. International Standard, March.",
"links": null
},
"BIBREF30": {
"ref_id": "b30",
"title": "Automatic thesaurus generation through multiple filtering",
"authors": [
{
"first": "K",
"middle": [],
"last": "Kageura",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Tsuji",
"suffix": ""
},
{
"first": "A",
"middle": [
"N"
],
"last": "Aizawa",
"suffix": ""
}
],
"year": 2000,
"venue": "Proceedings of the 18th Conference on Computational Linguistics",
"volume": "1",
"issue": "",
"pages": "397--403",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kageura, K., Tsuji, K., and Aizawa, A. N. (2000). Auto- matic thesaurus generation through multiple filtering. In Proceedings of the 18th Conference on Computational Linguistics -Volume 1, COLING '00, page 397-403, USA. Association for Computational Linguistics.",
"links": null
},
"BIBREF31": {
"ref_id": "b31",
"title": "Glossary of Key Information Security Terms. National Institute of Standards and Technology",
"authors": [
{
"first": "R",
"middle": [],
"last": "Kisserl",
"suffix": ""
}
],
"year": 2013,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kisserl, R., (2013). Glossary of Key Information Security Terms. National Institute of Standards and Technology, May. NISTIR 7298 Revision 2.",
"links": null
},
"BIBREF32": {
"ref_id": "b32",
"title": "Terminology systematization for Cybersecurity domain in Italian language",
"authors": [
{
"first": "C",
"middle": [],
"last": "Lanza",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Daille",
"suffix": ""
}
],
"year": 1991,
"venue": "TIA 2019 Terminologie et Intelligence Artificielle -Atelier TALN-RECITAL et IC (PFIA 2019)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lanza, C. and Daille, B. (2019). Terminology systematiza- tion for Cybersecurity domain in Italian language. In TIA 2019 Terminologie et Intelligence Artificielle -Atelier TALN-RECITAL et IC (PFIA 2019), Toulouse, France, July. Leech, G. (1991). The state of the art in corpus linguistics. Longman, London.",
"links": null
},
"BIBREF33": {
"ref_id": "b33",
"title": "Constitution d'une base bilingue de marqueurs de relations conceptuelles pour l'\u00e9laboration de ressources terminoontologiques",
"authors": [
{
"first": "L",
"middle": [],
"last": "Lefeuvre",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Condamines",
"suffix": ""
}
],
"year": 2015,
"venue": "Terminology and Artificial Intelligence (TIA'2015)",
"volume": "",
"issue": "",
"pages": "183--190",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lefeuvre, L. and Condamines, A. (2015). Constitution d'une base bilingue de marqueurs de relations con- ceptuelles pour l'\u00e9laboration de ressources termino- ontologiques. In Terminology and Artificial Intelligence (TIA'2015), pages 183-190, Granada, Spain.",
"links": null
},
"BIBREF34": {
"ref_id": "b34",
"title": "Analyse des marqueurs de relations conceptuelles en corpus sp\u00e9cialis\u00e9 : recensement, evaluation et caract\u00e9risation en fonction du domaine et du genre textuel",
"authors": [
{
"first": "L",
"middle": [],
"last": "Lefeuvre",
"suffix": ""
}
],
"year": 2017,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lefeuvre, L. (2017). Analyse des marqueurs de rela- tions conceptuelles en corpus sp\u00e9cialis\u00e9 : recensement, evaluation et caract\u00e9risation en fonction du domaine et du genre textuel. Ph.D. thesis. Th\u00e8se de doctorat Sci- ences du langage -U. Toulouse 2.",
"links": null
},
"BIBREF35": {
"ref_id": "b35",
"title": "Reference Lists for the Evaluation of Term Extraction Tools",
"authors": [
{
"first": "Loginova",
"middle": [],
"last": "Clouet",
"suffix": ""
},
{
"first": "E",
"middle": [],
"last": "Gojun",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Blancafort",
"suffix": ""
},
{
"first": "H",
"middle": [],
"last": "Guegan",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Gornostay",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "Heid",
"suffix": ""
},
{
"first": "U",
"middle": [],
"last": "",
"suffix": ""
}
],
"year": 2012,
"venue": "Terminology and Knowledge Engineering Conference (TKE)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Loginova Clouet, E., Gojun, A., Blancafort, H., Guegan, M., Gornostay, T., and Heid, U. (2012). Reference Lists for the Evaluation of Term Extraction Tools. In Termi- nology and Knowledge Engineering Conference (TKE), Madrid, Spain.",
"links": null
},
"BIBREF36": {
"ref_id": "b36",
"title": "The Corpus from a Terminographer's Viewpoint",
"authors": [
{
"first": "I",
"middle": [],
"last": "Meyer",
"suffix": ""
},
{
"first": "K",
"middle": [],
"last": "Mackintosh",
"suffix": ""
}
],
"year": 1996,
"venue": "International Journal of Corpus Linguistics",
"volume": "1",
"issue": "2",
"pages": "257--285",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Meyer, I. and Mackintosh, K. (1996). The Corpus from a Terminographer's Viewpoint. International Journal of Corpus Linguistics, 1(2):257-285.",
"links": null
},
"BIBREF37": {
"ref_id": "b37",
"title": "Linguistic regularities in continuous space word representations",
"authors": [
{
"first": "T",
"middle": [],
"last": "Mikolov",
"suffix": ""
},
{
"first": "S",
"middle": [
"W"
],
"last": "Yih",
"suffix": ""
},
{
"first": "G",
"middle": [],
"last": "Zweig",
"suffix": ""
}
],
"year": 2013,
"venue": "Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT-2013)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Mikolov, T., Yih, S. W.-t., and Zweig, G. (2013). Linguis- tic regularities in continuous space word representations. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL- HLT-2013). Association for Computational Linguistics, May.",
"links": null
},
"BIBREF38": {
"ref_id": "b38",
"title": "Projecting corpusbased semantic links on a thesaurus",
"authors": [
{
"first": "E",
"middle": [],
"last": "Morin",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Jacquemin",
"suffix": ""
}
],
"year": 1999,
"venue": "Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics",
"volume": "",
"issue": "",
"pages": "389--396",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Morin, E. and Jacquemin, C. (1999). Projecting corpus- based semantic links on a thesaurus. In Proceedings of the 37th Annual Meeting of the Association for Computa- tional Linguistics, page 389-396. Association for Com- putational Linguistics.",
"links": null
},
"BIBREF39": {
"ref_id": "b39",
"title": "Hierarchical embeddings for hypernymy detection and directionality",
"authors": [
{
"first": "K",
"middle": [
"A"
],
"last": "Nguyen",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "K\u00f6per",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Schulte Im Walde",
"suffix": ""
},
{
"first": "N",
"middle": [
"T"
],
"last": "Vu",
"suffix": ""
}
],
"year": 2017,
"venue": "Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing",
"volume": "",
"issue": "",
"pages": "233--243",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Nguyen, K. A., K\u00f6per, M., Schulte im Walde, S., and Vu, N. T. (2017). Hierarchical embeddings for hypernymy detection and directionality. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 233-243. Association for Computa- tional Linguistics.",
"links": null
},
"BIBREF40": {
"ref_id": "b40",
"title": "A framework for work task based thesaurus design",
"authors": [
{
"first": "M",
"middle": [
"L"
],
"last": "Nielsen",
"suffix": ""
}
],
"year": 2001,
"venue": "Journal of Documentation",
"volume": "57",
"issue": "6",
"pages": "774--797",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Nielsen, M. L. (2001). A framework for work task based thesaurus design. Journal of Documentation, 57(6):774- 797.",
"links": null
},
"BIBREF41": {
"ref_id": "b41",
"title": "Le th\u00e9saurus, un vocabulaire contr\u00f4l\u00e9 pour parler le m\u00eame langage",
"authors": [
{
"first": "M",
"middle": [],
"last": "Rennesson",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Georget",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Paillard",
"suffix": ""
},
{
"first": "O",
"middle": [],
"last": "Perrin",
"suffix": ""
},
{
"first": "H",
"middle": [],
"last": "Pigeotte",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "T\u00eate",
"suffix": ""
}
],
"year": 2020,
"venue": "M\u00e9decine Palliative",
"volume": "19",
"issue": "1",
"pages": "15--23",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Rennesson, M., Georget, M., Paillard, C., Perrin, O., Pi- geotte, H., and T\u00eate, C. (2020). Le th\u00e9saurus, un vocab- ulaire contr\u00f4l\u00e9 pour parler le m\u00eame langage. M\u00e9decine Palliative, 19(1):15 -23. Documentation et pratiques documentaires en soins palliatifsCoordonn\u00e9 par Caroline T\u00eate.",
"links": null
},
"BIBREF42": {
"ref_id": "b42",
"title": "Thesaurus construction guidelines: An introduction to thesauri and guidelines on their construction",
"authors": [
{
"first": "C",
"middle": [],
"last": "Ryan",
"suffix": ""
}
],
"year": 2014,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ryan, C. (2014). Thesaurus construction guidelines: An introduction to thesauri and guidelines on their construc- tion. Dublin: Royal Irish Academy and National Library of Ireland.",
"links": null
},
"BIBREF43": {
"ref_id": "b43",
"title": "Acquisition of semantic relations between terms: how far can we get with standard nlp tools?",
"authors": [
{
"first": "I",
"middle": [],
"last": "R\u00f6siger",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Bettinger",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Sch\u00e4fer",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Dorna",
"suffix": ""
},
{
"first": "U",
"middle": [],
"last": "Heid",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of COLING 2016: 5th International Workshop on Computational Terminology (CompuTerm)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "R\u00f6siger, I., Bettinger, J., Sch\u00e4fer, J., Dorna, M., and Heid, U. (2016). Acquisition of semantic relations between terms: how far can we get with standard nlp tools? In Proceedings of COLING 2016: 5th International Work- shop on Computational Terminology (CompuTerm), Os- aka, Japan.",
"links": null
},
"BIBREF44": {
"ref_id": "b44",
"title": "Poolparty: Skos thesaurus management utilizing linked data",
"authors": [
{
"first": "T",
"middle": [],
"last": "Schandl",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Blumauer",
"suffix": ""
}
],
"year": 2010,
"venue": "The Semantic Web: Research and Applications",
"volume": "",
"issue": "",
"pages": "421--425",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Schandl, T. and Blumauer, A. (2010). Poolparty: Skos thesaurus management utilizing linked data. In The Se- mantic Web: Research and Applications, pages 421-425, Berlin, Heidelberg. Springer Berlin Heidelberg.",
"links": null
},
"BIBREF45": {
"ref_id": "b45",
"title": "A Gold Standard for Multilingual Automatic Term Extraction from Comparable Corpora: Term Structure and Translation Equivalents",
"authors": [
{
"first": "A",
"middle": [
"R"
],
"last": "Terryn",
"suffix": ""
},
{
"first": "V",
"middle": [],
"last": "Hoste",
"suffix": ""
},
{
"first": "E",
"middle": [],
"last": "Lefever",
"suffix": ""
}
],
"year": 2018,
"venue": "Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Terryn, A. R., Hoste, V., and Lefever, E. (2018). A Gold Standard for Multilingual Automatic Term Extraction from Comparable Corpora: Term Structure and Trans- lation Equivalents. In Proceedings of the Eleventh Inter- national Conference on Language Resources and Eval- uation (LREC 2018), Miyazaki, Japan, May 7-12, 2018. European Language Resources Association (ELRA).",
"links": null
},
"BIBREF46": {
"ref_id": "b46",
"title": "Finding synonyms using automatic word alignment and measures of distributional similarity",
"authors": [
{
"first": "L",
"middle": [],
"last": "Van Der Plas",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Tiedemann",
"suffix": ""
}
],
"year": 2006,
"venue": "21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics ACL'06",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "van der Plas, L. and Tiedemann, J. (2006). Finding syn- onyms using automatic word alignment and measures of distributional similarity. In 21st International Con- ference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguis- tics ACL'06, Sydney, Australia.",
"links": null
},
"BIBREF47": {
"ref_id": "b47",
"title": "Simple methods for dealing with term variation and term alignment",
"authors": [
{
"first": "M",
"middle": [],
"last": "Weller",
"suffix": ""
},
{
"first": "A",
"middle": [],
"last": "Gojun",
"suffix": ""
},
{
"first": "U",
"middle": [],
"last": "Heid",
"suffix": ""
},
{
"first": "B",
"middle": [],
"last": "Daille",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Harastani",
"suffix": ""
}
],
"year": 2011,
"venue": "Proceedings of the 9th International Conference on Terminology and Artificial Intelligence",
"volume": "",
"issue": "",
"pages": "87--93",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Weller, M., Gojun, A., Heid, U., Daille, B., and Harastani, R. (2011). Simple methods for dealing with term varia- tion and term alignment. In Proceedings of the 9th Inter- national Conference on Terminology and Artificial Intel- ligence (TIA 2011), pages 87-93, Paris, France, Novem- ber. INALCO.",
"links": null
},
"BIBREF48": {
"ref_id": "b48",
"title": "Optimizing synonym extraction using monolingual and bilingual resources",
"authors": [
{
"first": "H",
"middle": [],
"last": "Wu",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Zhou",
"suffix": ""
}
],
"year": 2003,
"venue": "Proceedings of the second international workshop on Paraphrasing",
"volume": "16",
"issue": "",
"pages": "72--79",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Wu, H. and Zhou, M. (2003). Optimizing synonym ex- traction using monolingual and bilingual resources. In Proceedings of the second international workshop on Paraphrasing-Volume 16, pages 72-79. Association for Computational Linguistics.",
"links": null
},
"BIBREF49": {
"ref_id": "b49",
"title": "Automatic thesaurus construction",
"authors": [
{
"first": "D",
"middle": [],
"last": "Yang",
"suffix": ""
},
{
"first": "D",
"middle": [
"M"
],
"last": "Powers",
"suffix": ""
}
],
"year": 2008,
"venue": "Proceedings of the Thirty-First Australasian Conference on Computer Science",
"volume": "74",
"issue": "",
"pages": "147--156",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yang, D. and Powers, D. M. (2008a). Automatic the- saurus construction. In Proceedings of the Thirty-First Australasian Conference on Computer Science -Volume 74, ACSC '08, page 147-156, AUS. Australian Com- puter Society, Inc.",
"links": null
},
"BIBREF50": {
"ref_id": "b50",
"title": "Automatic thesaurus construction",
"authors": [
{
"first": "D",
"middle": [],
"last": "Yang",
"suffix": ""
},
{
"first": "D",
"middle": [
"M"
],
"last": "Powers",
"suffix": ""
}
],
"year": 2008,
"venue": "Proceedings of the Thirty-first Australasian Conference on Computer Science",
"volume": "74",
"issue": "",
"pages": "147--156",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yang, D. and Powers, D. M. (2008b). Automatic the- saurus construction. In Proceedings of the Thirty-first Australasian Conference on Computer Science -Volume 74, ACSC '08, pages 147-156, Darlinghurst, Australia, Australia. Australian Computer Society, Inc.",
"links": null
}
},
"ref_entries": {
"TABREF0": {
"num": null,
"html": null,
"content": "<table><tr><td>sum-</td></tr></table>",
"type_str": "table",
"text": ""
},
"TABREF1": {
"num": null,
"html": null,
"content": "<table/>",
"type_str": "table",
"text": "Number of words and documents of the Italian corpora: Clusit and Cybersecurity."
},
"TABREF2": {
"num": null,
"html": null,
"content": "<table><tr><td colspan=\"2\">Hyp Syn Rel Cause</td></tr><tr><td>#terms 172 63 110</td><td>68</td></tr><tr><td>#pairs 169 35 260</td><td>54</td></tr></table>",
"type_str": "table",
"text": "Size of the 5 term lists."
},
"TABREF3": {
"num": null,
"html": null,
"content": "<table/>",
"type_str": "table",
"text": "Semantic similarity evaluation list size. #terms indicates the total number of terms per semantic relation type, and #pairs indicates the number of pairs for each semantic relation."
},
"TABREF5": {
"num": null,
"html": null,
"content": "<table/>",
"type_str": "table",
"text": "Terminology extraction: number of candidate terms extracted by each tool for the Clusit and CyberSec corpora."
},
"TABREF6": {
"num": null,
"html": null,
"content": "<table><tr><td/><td/><td/><td/><td/><td/><td/><td/><td colspan=\"3\">Evaluation lists</td><td/><td/><td/><td/><td/></tr><tr><td/><td/><td/><td>Clusit</td><td/><td/><td>Glossary</td><td/><td/><td>Nist</td><td/><td/><td>Iso</td><td/><td/><td>Cyber</td></tr><tr><td>Corpus</td><td>coverage (%)</td><td/><td>100</td><td/><td/><td>36.2</td><td/><td/><td>22.3</td><td/><td/><td>55.6</td><td/><td/><td>49.3</td></tr><tr><td/><td>Tools</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td></tr><tr><td colspan=\"17\">T2K 0.19 33.1 0Clusit TermSuite 0.37 27.7 0.73 0.30 15.8 0.58 0.82 9.59 1.51 0.14 23.8 0.27 0.39 23.8 0.77</td></tr><tr><td/><td>PKE</td><td colspan=\"15\">0.85 69.8 1.68 0.35 20.4 0.69 0.95 12.4 1.76 0.17 32.9 0.34 0.46 30.7 0.91</td></tr><tr><td/><td>BERT</td><td colspan=\"15\">2.03 30.6 3.81 1.03 16.5 1.94 2.34 9.59 3.76 0.30 20.4 0.59 1.07 32.7 2.07</td></tr><tr><td/><td>coverage (%)</td><td/><td>61.3</td><td/><td/><td>72.5</td><td/><td/><td>35.3</td><td/><td/><td>67.0</td><td/><td/><td>100</td></tr><tr><td/><td>Tools</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td><td>P</td><td>R</td><td>F1</td></tr><tr><td/><td>T2K</td><td colspan=\"15\">0.01 23.7 0.02 0.02 42.2 0.04 0.05 21.2 0.10 0.01 47.7 0.02 0.01 30.7 0.02</td></tr><tr><td colspan=\"2\">CyberSec TermSuite</td><td colspan=\"15\">0.10 7.92 0.19 0.37 21.8 0.73 0.77 9.98 1.41 0.12 23.8 0.25 0.20 13.7 0.40</td></tr><tr><td/><td>PKE</td><td colspan=\"15\">0.05 49.0 0.10 0.06 44.0 0.12 0.12 21.2 0.24 0.02 44.3 0.04 0.05 46.5 0.10</td></tr><tr><td/><td>BERT</td><td colspan=\"15\">0.48 14.3 0.93 1.04 15.1 1.95 2.30 7.10 3.47 0.43 20.4 0.84 1.11 25.9 2.13</td></tr></table>",
"type_str": "table",
"text": ".38 0.18 21.4 0.36 0.48 13.1 0.93 0.09 37.5 0.18 0.15 21.4 0.32"
}
}
}
} |