File size: 74,101 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 |
{
"paper_id": "A00-2002",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T01:12:27.859937Z"
},
"title": "The Automatic Translation of Discourse Structures",
"authors": [
{
"first": "Daniel",
"middle": [],
"last": "Marcu",
"suffix": "",
"affiliation": {},
"email": "marcu@isi.edu"
},
{
"first": "Lynn",
"middle": [],
"last": "Carlson",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Ft",
"middle": [],
"last": "Meade",
"suffix": "",
"affiliation": {},
"email": ""
},
{
"first": "Maki",
"middle": [],
"last": "Watanabe",
"suffix": "",
"affiliation": {},
"email": "mwatanab@usc.edu"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "We empirically show that there are significant differences between the discourse structure of Japanese texts and the discourse structure of their corresponding English translations. To improve translation quality, we propose a computational model for rewriting discourse structures. When we train our model on a parallel corpus of manually built Japanese and English discourse structure trees, we learn to rewrite Japanese trees as trees that are closer to the natural English rendering than the original ones. 1 Motivation Almost all current MT systems process text one sentence at a time. Because of this limited focus, MT systems cannot regroup and reorder the clauses and sentences of an input text to achieve the most natural rendering in a target language. Yet, even between languages as close as English and French, there is a 10% mismatch in number of sentences-what is said in two sentences in one language is said in only one, or in three, in the other (Gale and Church, 1993). For distant language pairs, such as Japanese and English, the differences are more significant. Consider, for example, Japanese sentence (1), a word-byword \"gloss\" of it (2), and a two-sentence translation of it that was produced by a professional translator (3).",
"pdf_parse": {
"paper_id": "A00-2002",
"_pdf_hash": "",
"abstract": [
{
"text": "We empirically show that there are significant differences between the discourse structure of Japanese texts and the discourse structure of their corresponding English translations. To improve translation quality, we propose a computational model for rewriting discourse structures. When we train our model on a parallel corpus of manually built Japanese and English discourse structure trees, we learn to rewrite Japanese trees as trees that are closer to the natural English rendering than the original ones. 1 Motivation Almost all current MT systems process text one sentence at a time. Because of this limited focus, MT systems cannot regroup and reorder the clauses and sentences of an input text to achieve the most natural rendering in a target language. Yet, even between languages as close as English and French, there is a 10% mismatch in number of sentences-what is said in two sentences in one language is said in only one, or in three, in the other (Gale and Church, 1993). For distant language pairs, such as Japanese and English, the differences are more significant. Consider, for example, Japanese sentence (1), a word-byword \"gloss\" of it (2), and a two-sentence translation of it that was produced by a professional translator (3).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "(1) [ The labeled spans of text represent elementary discourse units (edus), i.e., minimal text spans that have an unambiguous discourse function (Mann and Thompson, 1988 ). If we analyze the text fragments closely, we will notice that in translating sentence (1), a professional translator chose to realize the information in Japanese unit 2 first (unit 2 in text (1) corresponds roughly to unit 1 in text (3)); to realize then some of the information in Japanese unit 1 (part of unit 1 in text (1) corresponds to unit 2 in text (3)); to fuse then information given in units 1, 3, and 5 in text (1) and realize it in English as unit 3; and so on. Also, the translator chose to repackage the information in the original Japanese sentence into two English sentences.",
"cite_spans": [
{
"start": 4,
"end": 5,
"text": "[",
"ref_id": null
},
{
"start": 146,
"end": 170,
"text": "(Mann and Thompson, 1988",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "At the elementary unit level, the correspondence between Japanese sentence (1) and its English translation (3) can be represented as in (4}, where j C e denotes the fact that the semantic content of unit j is realized fully in unit e; j D e denotes the fact that the semantic content of unit e is realized fully in unit j; j = e denotes the fact that units j and e are semantically equivalent; and j ~ e denotes the fact that there is a semantic overlap between units j and e, but neither proper inclusion nor proper equivalence.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": ".!1 D e2;jt -~ e3;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": ".12 ----el; 33 C e3;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": ".14 ~ e4;j4 ~ es; .15 ~ e3;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": ".16 C e6;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": ".17 C e6 Hence. the mappings in (4) provide all explicit representation of the way information is re-ordered and re-packaged when translated from Japanese into English. However, when translating text, it is also the case that t he rhetorical rendering changes. What is realized ill Japanese using an CONTRAST relation can be realized in English using, for example, a COXl- PARISON or a CONCESSION relation.",
"cite_spans": [
{
"start": 373,
"end": 380,
"text": "PARISON",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "Figure I presents in the style of Mann and Thompson (1988) the discourse structures of text fragments (1) and (3), Each discourse structure is a tree whose leaves correspond to the edus and whose internal nodes correspond to contiguous text spans. Each node is characterized by a status (NUCLEUS or SATELLITE) and a rhetorical relation, which is a relation that holds between two non-overlapping text spans. The distinction between nuclei and satellites comes from the empirical observation that the nucleus expresses what is more essential to the writer's intention than the satellite: and that the nucleus of a rhetorical relation is comprehensible independent of tile satellite, but not vice versa. When spans are equally important, the relation is nmltinuclear: for example, the CONTRAST relation that holds between unit [3] and span [4.5] in the rhetorical structure of the English text in figure 1 is nmhinuclear. Rhetorical relations that end in the suffix \"'-e'\" denote relations that correspond to embedded syntactic con-stituents. For example, the ELABORATION-OBJECT-ATTRIBUTE-E relation that holds between units 2 and 1 in the English discourse structure corresponds to a restrictive relative.",
"cite_spans": [
{
"start": 838,
"end": 843,
"text": "[4.5]",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "If one knows the mappings at the edu level, one can determine the mappings at the span (discourse constituent) level as well. For example, using the elementary mappings in (4), one call determine that Japanese span [1,2] corresponds to English span [I,2] , Japanese unit [4] to English span [4, 5] , Japanese span [6.7] to English unit [6], Japanese span [1.5] to English span [1.5], and so on. As Figure 1 shows, the CONCESSION relation that holds between spans [1, 5] and [6, 7] in the Japanese tree corresponds to a similar relation that. holds between span [1,5] and unit [6] in the English tree (modulo the fact that, in Japanese, the relation holds between sent ence fragments, while in English it holds between full sentences). However, the TEMPORAL-AFTER relation that holds between units [:3] and [4] ill the Japanese tree is realized as a CONTRAST relation between unit [3] and span [4.5] in MT systems at the syntactic level. For example, the re-ordering of units 1 and 2, can be dealt with using only syntactic models. However, as we will see in Section 2, there are significant differences between Japanese and English with respect to the way information is packaged and organized rhetorically not only at the sentence level, but also, at the paragraph and text levels. More specifically, as humans translate Japanese into English, they re-order the clauses, sentences, and paragraphs of Japanese texts, they re-package the information into clauses, sentences, and paragraphs that are not a one-to-one mapping of the original Japanese units, and they rhetorically re-organize the structure of the translated text so as to reflect rhetorical constraints specific to English. If a translation system is to produce text that is not only grammatical but also coherent, it will have to ensure that the discourse structure of the target text reflects the natural renderings of the target language, and not that of the source language. In Section 2, we empirically show that there are significant differences between the rhetorical structure of Japanese texts and their corresponding English translations. These differences justify our investigation into developing computational models for discourse structure rewriting. In Section 3, we present such a rewriting model, which re-orders the edus of the original text, determines English-specific clause, sentence, and paragraph boundaries, and rebuilds the Japanese discourse structure of a text using English-specific rhetorical renderings. In Section 4, we evaluate the performance of an implementation of this model. We end with a discussion.",
"cite_spans": [
{
"start": 249,
"end": 254,
"text": "[I,2]",
"ref_id": null
},
{
"start": 291,
"end": 294,
"text": "[4,",
"ref_id": null
},
{
"start": 295,
"end": 297,
"text": "5]",
"ref_id": null
},
{
"start": 463,
"end": 466,
"text": "[1,",
"ref_id": null
},
{
"start": 467,
"end": 469,
"text": "5]",
"ref_id": null
},
{
"start": 474,
"end": 477,
"text": "[6,",
"ref_id": null
},
{
"start": 478,
"end": 480,
"text": "7]",
"ref_id": null
},
{
"start": 576,
"end": 579,
"text": "[6]",
"ref_id": null
},
{
"start": 893,
"end": 898,
"text": "[4.5]",
"ref_id": null
}
],
"ref_spans": [
{
"start": 395,
"end": 406,
"text": "As Figure 1",
"ref_id": "FIGREF1"
}
],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "In order to assess the role of discourse structure in MT, we built manually a corpus of discourse trees for 40 Japanese texts and their corresponding translations. The texts were selected randomly from the ARPA corpus (White and O'Connell, 1994) . On average, each text had about 460 words. The Japanese texts had a total of 335 paragraphs and 773 sentences. The English texts had a total of 337 paragraphs and 827 sentences.",
"cite_spans": [
{
"start": 218,
"end": 245,
"text": "(White and O'Connell, 1994)",
"ref_id": "BIBREF12"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "We developed a discourse annotation protocol for Japanese and English along the lines followed by . We used Marcu's discourse annotation tool (1999) in order to manually construct the discourse structure of all Japanese and English texts in the corpus. 10% of the Japanese and English texts were rhetorically labeled by two of us.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "The tool and the annotation protocol are available at http://www.isi.edu/~marcu/software/. The annotation procedure yielded over the entire corpus 2641 Japanese edus and 2363 English edus.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "We computed the reliability of the annotation using 's method for computing kappa statistics (Siegel and Castellan, 1988) over hierarchical structures. Table 1 displays average kappa statistics that reflect the reliability of the annotation of elementary discourse units, k~,, hierarchical discourse spans, ks, hierarchical nuclearity assignments, k,~, and hierarchical rhetorical relation assignments, k~. Kappa figures higher than 0.8 correspond to good agreement; kappa figures higher than 0.6 correspond to acceptable agreement. All kappa statistics were statistically significant at levels higher than a = 0.01. In addition to the kappa statistics, table 1 also displays in parentheses the average number of data points per document, over which the kappa statistics were computed.",
"cite_spans": [
{
"start": 93,
"end": 121,
"text": "(Siegel and Castellan, 1988)",
"ref_id": "BIBREF9"
}
],
"ref_spans": [
{
"start": 152,
"end": 159,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "For each pair of Japanese-English discourse structures, we also built manually an alignment file, which specified in the notation discussed on page 1 the correspondence between the edus of the Japanese text and the edus of its English translation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "We computed the similarity between English and Japanese discourse trees using labeled recall and precision figures that reflected the resemblance of tile Japanese and English discourse structures with respect to their assignment of edu boundaries, hierarchical spans, nuclearity, and rhetorical relations.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "Because the trees we compared differ from one language to the other in the number of elementary units, the order of these units, and the way the units are grouped recursively into discourse spans, we computed two types of recall and precision figures. In computing Position-Dependent (P-D) recall and precision figures, a Japanese span was considered to match an English span when the Japanese span contained all the Japanese edus that corresponded to tile edus in the English span, and when the Japanese and English spans appeared in tile same position with respect to the overall structure. For example, the English tree in figure 1 is characterized by 10 subsentential spans: [3, 5] , and [1, 5] . (Span [1,6] subsumes 2 sentences, so it is not sub-sentential.) The Japanese discourse tree has only 4 spans that could be matched in the same positions with English spans, namely spans [1,2], [4], [5] , and [1, 5] . Hence the similarity between the Japanese tree and the English tree with respect Level Units Spans Status/Nuclearity Relations P-D P P-D R P-D P P-DR P-DP P-DR P-I) P P-DR Table 2 : Similarity of the Japanese to their discourse structure below the sentence level has a recall of 4/10 and a precision of 4/11 (in Figure 1, there are 11 sub-sentential Japanese spans).",
"cite_spans": [
{
"start": 679,
"end": 682,
"text": "[3,",
"ref_id": null
},
{
"start": 683,
"end": 685,
"text": "5]",
"ref_id": null
},
{
"start": 692,
"end": 695,
"text": "[1,",
"ref_id": null
},
{
"start": 696,
"end": 698,
"text": "5]",
"ref_id": null
},
{
"start": 899,
"end": 902,
"text": "[5]",
"ref_id": null
},
{
"start": 909,
"end": 912,
"text": "[1,",
"ref_id": null
},
{
"start": 913,
"end": 915,
"text": "5]",
"ref_id": null
}
],
"ref_spans": [
{
"start": 1090,
"end": 1097,
"text": "Table 2",
"ref_id": null
},
{
"start": 1230,
"end": 1236,
"text": "Figure",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "[1], [2], [3], [4], [5], [6], [1,2], [4,5],",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "In computing Position-Independent (P-I) recall and precision figures, even when a Japanese span \"floated\" during the translation to a position in the English tree that was different from the position in the initial tree, the P-I recall and precision figures were not affected. The Position-Independent figures reflect the intuition that if two trees tl and t2 both have a subtree t, tl and t2 are more similar than if they were if they didn't share any tree. At the sentence level, we hence assume that if, for example, the syntactic structure of a relative clause is translated appropriately (even though it is not appropriately attached), this is better than translating wrongly that clause. The Position-Independent figures offer a more optimistic metric for comparing discourse trees. They span a wider range of values than the Position-Dependent figures, which enable a finer grained comparison, which in turn enables a better characterization of the differences between Japanese and English discourse structures. When one takes an optimistic stance, for the spans at the sub-sentential level in the trees in Table 1 In order to provide a better estimate of how close two discourse trees were, we computed Position-Dependent and -Independent recall and precision figures for the sentential level (where units are given by edus and spans are given by sets of edus or single sentences); paragraph level (where units are given by sentences and spans are given by sets of sentences or single paragraphs); and text level (where units are given by paragraphs and spans are given by sets of paragraphs). These figures offer a detailed picture of how discourse structures and relations are mapped from one language to the other across all and English discourse structures discourse levels, from sentence to text. The differences at the sentence level can be explained by differences between the syntactic structures of Japanese and English. The differences at the paragraph and text levels have a purely rhetorical explanation.",
"cite_spans": [],
"ref_spans": [
{
"start": 1114,
"end": 1121,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "As expected, when we computed the recall and precision figures with respect to the nuclearity and relation assignments, we also factored in the statuses and the rhetorical relations that labeled each pair of spans. Table 2 smnmarizes the results (P-D and P-I (R)ecall and (P)recision figures) for each level (Sentence, Paragraph, and Text). The numbers in the \"Weighted Average\" line report averages of the Sentence-, Paragraph-, and Text-specific figures, weighted according to the number of units at each level. The numbers in the \"All\" line reflect recall and precision figures computed across the entire trees, with no attention paid to sentence and paragraph boundaries.",
"cite_spans": [],
"ref_spans": [
{
"start": 215,
"end": 222,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "Given the significantly different syntactic structures of Japanese and English, we were not surprised by the low recall and precision results that reflect the similarity between discourse trees built below the sentence level. However, as Table 2 shows, there are significant differences between discourse trees at the paragraph and text levels as well. For exampie, the Position-Independent figures show that only about 62% of the sentences and only about 53% of the hierarchical spans built across sentences could be matched between the two corpora. When one looks at the status and rhetorical relations associated with the spans built across sentences at the paragraph level, the P-I recall and precision figures drop to about 43% and 35% respectively.",
"cite_spans": [],
"ref_spans": [
{
"start": 238,
"end": 245,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "The differences in recall and precision are explained both by differences in the way information is packaged into paragraphs in the two languages and the way it is structured rhetorically both within and above the paragraph level.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "These results strongly suggest that if one attempts to translate Japanese into English on a sentence-bysentence basis, it is likely that the resulting text will be unnatural from a discourse perspective. For example, if some information rendered using a CON-TRAST relation in Japanese is rendered using an ELABORATION relation in English, it would be inappropriate to use a discourse marker like \"but\" in the English translation, although that would be consistent with the Japanese discourse structure. An inspection of the rhetorical mappings between Japanese and English revealed that some Japanese rhetorical renderings are consistently mapped into one or a few preferred renderings in English. For example, 34 of 115 CONTRAST relations in the Japanese texts are mapped into CONTRAST relations in English; 27 become nuclei of relations such as ANTITHE-SIS and CONCESSION, 14 are translated as COMPAR-ISON relations, 6 as satellites of CONCESSION relations, 5 as LIST relations, etc. Our goal is to learn these systematic discourse mapping rules and exploit them in a machine translation context.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experiment",
"sec_num": "2"
},
{
"text": "Towards a discourse-based machine translation system",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "3",
"sec_num": null
},
{
"text": "We are currently working towards building the modules of a Discourse-Based Machine Translation system that works along the following lines.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Overall architecture",
"sec_num": "3.1"
},
{
"text": "1. A discourse parser, such as those described by Sumita et al. (1992) , Kurohashi (1994) , and MarcH (1999), initially derives the discourse structure of the text given as input.",
"cite_spans": [
{
"start": 50,
"end": 70,
"text": "Sumita et al. (1992)",
"ref_id": "BIBREF11"
},
{
"start": 73,
"end": 89,
"text": "Kurohashi (1994)",
"ref_id": "BIBREF2"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Overall architecture",
"sec_num": "3.1"
},
{
"text": "A discourse-structure transfer module rewrites the discourse structure of the input text so as to reflect a discourse rendering that is natural to the target language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "2.",
"sec_num": null
},
{
"text": "into the target language using translation and language models that incorporate discoursespecific features, which are extracted from the outputs of the discourse parser and discourse transfer modules.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "A statistical module maps the input text",
"sec_num": "3."
},
{
"text": "In this paper, we focus only on the discoursestructure transfer module. That is, we investigate the feasibility of building such a module.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "A statistical module maps the input text",
"sec_num": "3."
},
{
"text": "In order to learn to rewrite discourse structure trees, we first address a related problem, which we define below:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "Definition 3.1 Given two trees Ts and Tt and a correspondence Table C defined between Ts and Tt at the leaf level in terms of-----, C, D, and ~ relations, find a sequence of actions that rewrites the tree T~ into Tt.",
"cite_spans": [],
"ref_spans": [
{
"start": 62,
"end": 69,
"text": "Table C",
"ref_id": null
}
],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "If for any tuple (Ts, Tt, C> such a sequence of actions can be derived, it is then possible to use a corpus of (Ts, Tt, C) tuples in order to automatically learn to derive from an unseen tree Ts,, which has the same structural properties as the trees Ts, a tree Ttj, which has structural properties similar to those of the trees Tt. In order to solve the problem in definition 3.1, we extend the shift-reduce parsing paradigm applied by Magerman (1995) , Hermjakob and Mooney (1997), and MarcH (1999) . In this extended paradigm, the transfer process starts with an empty Stack and an Input List that contains a sequence of elementary discourse trees edts, one edt for each edu in the tree Ts given as input. The status and rhetorical relation associated with each edt is undefined. At each step, the transfer module applies an operation that is aimed at building from the units in T, the discourse tree Tt. In the context of our discourse-transfer module, we need 7 types of operations:",
"cite_spans": [
{
"start": 437,
"end": 452,
"text": "Magerman (1995)",
"ref_id": "BIBREF3"
},
{
"start": 455,
"end": 468,
"text": "Hermjakob and",
"ref_id": "BIBREF1"
},
{
"start": 469,
"end": 487,
"text": "Mooney (1997), and",
"ref_id": "BIBREF1"
},
{
"start": 488,
"end": 500,
"text": "MarcH (1999)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 SHIFT operations transfer the first edt from the input list into the stack;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 REDUCE operations pop the two discourse trees located at the top of the stack; combine them into a new tree updating the statuses and rhetorical relation names of the trees involved in the operation; and push the new tree on the top of the stack. These operations are used to build the structure of the discourse tree in the target language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 BREAK operations are used in order to break the edt at the beginning of the input list into a predetermined number of units. These operations are used to ensure that the resulting tree has the same number of edts as Tt.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "A BREAK operation is necessary whenever a Japanese edu is mapped into nmltiple English units.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 CREATE-NEXT operations are used in order to create English discourse constituents that have no correspondent in the Japanese tree.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 FUSE operations are used in order to fuse the edt at the top of the stack into the tree that immediately precedes it. These operations are used whenever multiple Japanese edus are mapped into one English edu.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 SWAP operations swap the edt at the beginning of the input list with an edt found one or more positions to the right. These operations are necessary for re-ordering discourse constituents.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 ASSIGNTYPE operations assign one or more of the following types to the tree at the top of the stack: Unit, MultiUnit, Sentence, Paragraph, MultiParagraph, and Text. These op-erations are necessary in order to ensure sentence and paragraph boundaries that are specific to the target language.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "For example, the first sentence of the English tree in Figure 1 can be obtained from the original Japanese sequence by following the sequence of actions (5), whose effects are shown in Figure 2 . For the purpose of compactness, the figure does not illustrate the effect of ASSIGNTYPE actions. For the same purpose, some lines correspond to more than one action, For our corpus, in order to enable a discoursebased transfer module to derive any English discourse tree starting from any Japanese discourse tree, it is sufficient to implement: * one SHIFT operation;",
"cite_spans": [],
"ref_spans": [
{
"start": 55,
"end": 63,
"text": "Figure 1",
"ref_id": "FIGREF1"
},
{
"start": 185,
"end": 193,
"text": "Figure 2",
"ref_id": "FIGREF3"
}
],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 3 x 2 \u00d7 85 REDUCE operations; (For each of the three possible pairs of nuclearity assignments NUCLEUS-SATELLITE (NS), SATELLITE-NUCLEUS (SN), AND NUCLEUS-NUCLEUS (NN), there are two possible ways to reduce two adjacent trees (one results in a binary tree, the other in a non-binary tree (Marcu, 1999) ), and 85 relation names.)",
"cite_spans": [
{
"start": 289,
"end": 302,
"text": "(Marcu, 1999)",
"ref_id": "BIBREF5"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 three types of BREAK operations; (In our corpus, a Japanese unit is broken into two, three, or at most four units.)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 one type of CREATE-NEXT operation;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 one type of FUSE operation;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 eleven types of SWAP operations; (In our corpus, Japanese units are at most l l positions away from their location in an Englishspecific rendering.)",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "\u2022 seven types of ASSIGN~]~YPE operations: Unit, MultiUnit, Sentence, MultiSentence, Paragraph, MultiParagraph, and Text.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "These actions are sufficient for rewriting any tree Ts into any tree Tt, where Tt may have a different number of edus, where the edus of Tt may have a different ordering than the edus of Ts, and where the hierarchical structures of the two trees may be different as well.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "3.3 Learning the parameters of the discourse-transfer model We associate with each configuration of our transfer model a learning case. The cases were generated by a program that automatically derived the sequence of actions that mapped the Japanese trees in our corpus into the sibling English trees, using the correspondences at the elementary unit level that were constructed manually. Overall, the 40 pairs of Japanese and English discourse trees yielded 14108 cases.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "To each learning example, we associated a set of features from the following classes:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "Operational and discourse features reflect the number of trees in the stack, the input list, and the types of the last five operations. They encode information pertaining to the types of the partial trees built up to a certain time and the rhetorical relations that hold between these trees.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "Correspondence-based features reflect the nuclearity, rhetorical relations, and types of the Japanese trees that correspond to the English-like partial trees derived up to a given time.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "Lexicalfeatures specify whether the Japanese spans that correspond to the structures derived up to a given time use potential discourse markers, such as dakara (because) and no ni (although).",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "The discourse transfer module uses the C4.5 program (Quinlan, 1993) in order to learn decision trees and rules that specify how Japanese discourse trees should be mapped into English-like trees. A ten-fold cross-validation evaluation of the classifier yielded an accuracy of 70.2% (+ 0.21).",
"cite_spans": [
{
"start": 52,
"end": 67,
"text": "(Quinlan, 1993)",
"ref_id": "BIBREF8"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "In order to better understand the strengths and weaknesses of the classifier, we also attempted to break the problem into smaller components. Hence, instead of learning all actions at once, we attempted to learn first whether the rewriting procedure should choose a SHIFT, REDUCE, BREAK, FUSE, SWAP, or ASSIGNTYPE operation (the \"Main Action Type\" classifier in table 3), and only then to refine this decision by determining what type of reduce operation to perform, how many units to break a Japanese units into, how big the distance to the SWAP-ed unit should be, and what type of ASSIGNTYPE operation one should perform. Table 3 shows the sizes of each STACK 2 2 1\"",
"cite_spans": [],
"ref_spans": [
{
"start": 624,
"end": 670,
"text": "Table 3 shows the sizes of each STACK 2 2",
"ref_id": "TABREF5"
}
],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "[~A BORATION_(IB~TI- data set and the performance of each of these classitiers, as determined using a ten-fold cross-validation procedure. For the purpose of comparison, each classifier is paired with a majority baseline.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "The results in Table 3 show that the most difficult subtasks to learn are that of determining the number of units a Japanese unit should be broken into and that of determining the distance to the unit that is to be swapped. The features we used are not able to refine the baseline classifiers for these action types. The confusion matrix for the \"Main Action Type\" classifier (see Table 5 ) shows that the system has trouble mostly identifying BREAK and CREATE-NEXT actions. The system has difficulty learning what type of nuclearity ordering to prefer (the \"Nuclearity-Reduce\" classifier) and what re-lation to choose for the English-like structure (the \"Relation-Reduce\" classifier). Figure 3 shows a typical learning curve, the one that corresponds to the \"Reduce Relation\" classifier. Our learning curves suggest that more training data may improve performance. However, they also suggest that better features may be needed in order to improve performance significantly. Table 4 displays some learned rules. The first rule accounts for rhetorical mappings in which the order of the nucleus and satellite of an ATTRIBUTION relation is changed when translated from Japanese into English. The second rule was learned in order to map EXAMPLE Japanese satellites into EVIDENCE English satellites. ",
"cite_spans": [],
"ref_spans": [
{
"start": 15,
"end": 22,
"text": "Table 3",
"ref_id": "TABREF5"
},
{
"start": 381,
"end": 388,
"text": "Table 5",
"ref_id": "TABREF8"
},
{
"start": 686,
"end": 694,
"text": "Figure 3",
"ref_id": "FIGREF4"
},
{
"start": 975,
"end": 982,
"text": "Table 4",
"ref_id": "TABREF6"
}
],
"eq_spans": [],
"section": "The discourse-based transfer model",
"sec_num": "3.2"
},
{
"text": "By applying the General classifier or the other six classifiers successively, one can map any Japanese discourse tree into a tree whose structure comes closer to the natural rendering of English. To evaluate the discourse-based transfer module, we carried out a ten-fold cross-validation experiment. That is, we trained the classifiers on 36 pairs of manually built and aligned discourse structures, and we then used the learned classifiers in order to map 4 unseen Japanese discourse trees into English-like trees. We measured the similarity of the derived trees with the English trees built manually, using the metrics discussed in Section 2. We repeated the procedure ten times, each time training and testing on different subsets of tree pairs. We take the results reported in Table 2 as a baseline for our model. The baseline corresponds to applying no knowledge of discourse. Table 6 displays the absolute improvement (in percentage points) in recall and precision figures obtained when the General classifier was used to map Japanese trees into English-looking trees. The General classifier yielded the best results. The results in Table 6 are averaged over a ten-fold cross-validation experiment.",
"cite_spans": [],
"ref_spans": [
{
"start": 781,
"end": 788,
"text": "Table 2",
"ref_id": null
},
{
"start": 882,
"end": 889,
"text": "Table 6",
"ref_id": null
},
{
"start": 1139,
"end": 1146,
"text": "Table 6",
"ref_id": null
}
],
"eq_spans": [],
"section": "Evaluation of the discourse-based transfer module",
"sec_num": "4"
},
{
"text": "The results in Table 6 show that our model outperforms the baseline with respect to building English-like discourse structures for sentences, but it under-performs the baseline with respect to building English-like structures at the paragraph and text levels. The main shortcoming of our model seems to come from its low performance in assigning paragraph boundaries. Because our classifier does not learn correctly which spans to consider paragraphs and which spans not, the recall and precision results at the paragraph and text levels are negatively affected. The poorer results at the paragraph and text levels can be also explained by errors whose effect cumulates during the step-by-step tree-reconstruction procedure; and by the fact that, for these levels, there is less data to learn from.",
"cite_spans": [],
"ref_spans": [
{
"start": 15,
"end": 22,
"text": "Table 6",
"ref_id": null
}
],
"eq_spans": [],
"section": "Evaluation of the discourse-based transfer module",
"sec_num": "4"
},
{
"text": "However, if one ignores the sentence and paragraph boundaries and evaluates the discourse structures overall, one can see that our model outperforms the baseline on all accounts according to the Position-Dependent evaluation; outperforms the baseline with respect to the assignment of elementary units, hierarchical spans, and nuclearity statuses according to the Position-Independent evaluation and under-performs the baseline only slightly +9.1 +25.5 +2.0 +19.9 +0.4 +13.4 -0.01 +8.4 Paragraph -14.7 +1.4 -12.5 -1.7 -11.0 -2.4 -9.9 -3.3 Text -9.6 -13.5 -7.1 -11.1 -6.3 -10.0 -5.2 -8.8 Weighted Average +1.5 +14.1 -2.1 +9.9 -3.1 +6.4 -3.0 +3.9 All -1.2 +2.5 -0.1 +2.9 +0.6 +3.5 +0.7 +2.6 P-I R P-I P P-I R P-I P P-I R P-I P Table 6 : Relative evaluation of the discourse-based transfer module with respect to the figures in Table 2. with respect to the rhetorical relation assignment according to the Position-Independent evaluation. More sophisticated discourse features, such as those discussed by Maynard (1998), for example, and a tighter integration with the lexicogrammar of the two languages may yield better cues for learning discourse-based translation models.",
"cite_spans": [],
"ref_spans": [
{
"start": 725,
"end": 732,
"text": "Table 6",
"ref_id": null
},
{
"start": 825,
"end": 833,
"text": "Table 2.",
"ref_id": null
}
],
"eq_spans": [],
"section": "Evaluation of the discourse-based transfer module",
"sec_num": "4"
},
{
"text": "We presented a systematic empirical study of the role of discourse structure in MT. Our study strongly supports the need for enriching MT systems with a discourse module, capable of re-ordering and repackaging the information in a source text in a way that is consistent with the discourse rendering of a target language. We presented an extended shiftreduce parsing model that can be used to map discourse trees specific to a source language into discourse trees specific to a target language. Our model outperforms a baseline with respect to its ability to predict the discourse structure of sentences. Our model also outperforms the baseline with respect to its ability to derive discourse structures that are closer to the natural, rhetorical rendering in a target language than the original discourse structures in the source language. Our model is still unable to determine correctly how to re-package sentences into paragraphs; a better understanding of the notion of \"paragraph\" is required in order to improve this.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "5"
}
],
"back_matter": [],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "A program for aligning sentences in bilingual corpora",
"authors": [
{
"first": "A",
"middle": [],
"last": "William",
"suffix": ""
},
{
"first": "Kenneth",
"middle": [
"W"
],
"last": "Gale",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Church",
"suffix": ""
}
],
"year": 1993,
"venue": "Computational Linguistics",
"volume": "19",
"issue": "1",
"pages": "75--102",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "William A. Gale and Kenneth W. Church. 1993. A program for aligning sentences in bilingual cor- pora. Computational Linguistics, 19(1):75-102.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Learning parse and translation decisions from examples with rich context",
"authors": [
{
"first": "Ulf",
"middle": [],
"last": "Hermjakob",
"suffix": ""
},
{
"first": "Raymond",
"middle": [
"J"
],
"last": "Mooney",
"suffix": ""
}
],
"year": 1997,
"venue": "Proc. of ACL'97",
"volume": "",
"issue": "",
"pages": "482--489",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Ulf Hermjakob and Raymond J. Mooney. 1997. Learning parse and translation decisions from ex- amples with rich context. In Proc. of ACL'97, pages 482-489, Madrid, Spain..",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Automatic detection of discourse structure by checking surface information in sentences",
"authors": [
{
"first": "Sadao",
"middle": [],
"last": "Kurohashi",
"suffix": ""
},
{
"first": "Makoto",
"middle": [],
"last": "Nagao",
"suffix": ""
}
],
"year": 1994,
"venue": "Proc. of COLING'94",
"volume": "2",
"issue": "",
"pages": "1123--1127",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sadao Kurohashi and Makoto Nagao. 1994. Auto- matic detection of discourse structure by check- ing surface information in sentences. In Proc. of COLING'94, volume 2, pages 1123-1127, Kyoto, Japan.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Statistical decision-tree models for parsing",
"authors": [
{
"first": "David",
"middle": [
"M"
],
"last": "Magerman",
"suffix": ""
}
],
"year": 1995,
"venue": "Proc. of A CL '95",
"volume": "",
"issue": "",
"pages": "276--283",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "David M. Magerman. 1995. Statistical decision-tree models for parsing. In Proc. of A CL '95, pages 276-283, Cambridge, Massachusetts.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Rhetorical structure theory: Toward a functional theory of text organization",
"authors": [
{
"first": "C",
"middle": [],
"last": "William",
"suffix": ""
},
{
"first": "Sandra",
"middle": [
"A"
],
"last": "Mann",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Thompson",
"suffix": ""
}
],
"year": 1988,
"venue": "Text",
"volume": "8",
"issue": "3",
"pages": "243--281",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "William C. Mann and Sandra A. Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243-281.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "A decision-based approach to rhetorical parsing",
"authors": [
{
"first": "Daniel",
"middle": [],
"last": "Marcu",
"suffix": ""
}
],
"year": 1999,
"venue": "Proc. of A CL'99",
"volume": "",
"issue": "",
"pages": "365--372",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Daniel Marcu. 1999. A decision-based approach to rhetorical parsing. In Proc. of A CL'99, pages 365- 372, Maryland.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "Experiments in constructing a corpus of discourse trees",
"authors": [
{
"first": "Daniel",
"middle": [],
"last": "Marcu",
"suffix": ""
},
{
"first": "Estibaliz",
"middle": [],
"last": "Amorrortu",
"suffix": ""
},
{
"first": "Magdalena",
"middle": [],
"last": "Romera",
"suffix": ""
}
],
"year": 1999,
"venue": "Proc. of the A CL'99 Workshop on Standards and Tools for Discourse Tagging",
"volume": "",
"issue": "",
"pages": "48--57",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Daniel Marcu, Estibaliz Amorrortu, and Magdalena Romera. 1999. Experiments in constructing a cor- pus of discourse trees. In Proc. of the A CL'99 Workshop on Standards and Tools for Discourse Tagging, pages 48-57, Maryland.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Principles of Japanese Discourse: A Handbook",
"authors": [
{
"first": "K",
"middle": [],
"last": "Senko",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Maynard",
"suffix": ""
}
],
"year": 1998,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Senko K. Maynard. 1998. Principles of Japanese Discourse: A Handbook. Cambridge Univ. Press.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "C4.5: Programs for Machine Learning",
"authors": [
{
"first": "J",
"middle": [],
"last": "",
"suffix": ""
},
{
"first": "Ross",
"middle": [],
"last": "Quinlan",
"suffix": ""
}
],
"year": 1993,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. Ross Quinlan. 1993. C4.5: Programs for Machine Learning. Morgan Kaufmann Publishers.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Nonparametric Statistics for the Behavioral Sciences",
"authors": [
{
"first": "Sidney",
"middle": [],
"last": "Siegel",
"suffix": ""
},
{
"first": "N",
"middle": [
"J"
],
"last": "Castellan",
"suffix": ""
}
],
"year": 1988,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sidney Siegel and N.J. Castellan. 1988. Non- parametric Statistics for the Behavioral Sciences.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "A discourse structure analyzer for Japanese text",
"authors": [
{
"first": "Kazuo",
"middle": [],
"last": "Sumita",
"suffix": ""
},
{
"first": "Kenji",
"middle": [],
"last": "Ono",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "Chino",
"suffix": ""
},
{
"first": "Teruhiko",
"middle": [],
"last": "Ukita",
"suffix": ""
},
{
"first": "Shin'ya",
"middle": [],
"last": "Amano",
"suffix": ""
}
],
"year": 1992,
"venue": "Proceedings of the International Conference on Fifth Generation Computer Systems",
"volume": "",
"issue": "",
"pages": "1133--1140",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Kazuo Sumita, Kenji Ono, T. Chino, Teruhiko Ukita, and Shin'ya Amano. 1992. A discourse structure analyzer for Japanese text. In Proceed- ings of the International Conference on Fifth Gen- eration Computer Systems, v 2, pages 1133-1140.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Evaluation in the ARPA machine-translation program: 1993 methodology",
"authors": [
{
"first": "J",
"middle": [],
"last": "White",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "O'connell",
"suffix": ""
}
],
"year": 1994,
"venue": "Proceedings of the ARPA Human Language Technology Workshop",
"volume": "",
"issue": "",
"pages": "135--140",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "J. White and T. O'Connell. 1994. Evaluation in the ARPA machine-translation program: 1993 methodology. In Proceedings of the ARPA Human Language Technology Workshop, pages 135-140, Washington, D.C.",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"text": "................................................................................................................. ==:~::::\u00b02'.~2~?:::-~.:.,~a .................",
"type_str": "figure",
"uris": null,
"num": null
},
"FIGREF1": {
"text": "The discourse structures of texts (1) and (3).",
"type_str": "figure",
"uris": null,
"num": null
},
"FIGREF3": {
"text": "Example of incremental tree reconstruction.",
"type_str": "figure",
"uris": null,
"num": null
},
"FIGREF4": {
"text": "Learning curve for the Relation-Reduce classifier. if rhetRelOfStack-llnJapTree = ATTRIBUTION then rhetRelOffFopStacklnEngTree ~ ATTRIBUTION if rhetRelOffFopStacklnJapTree ----EXAMPLE A isSentenceTheLastUnitlnJapTreeOfropStack = false then rhetRelOfI'opStackInEngTree ~ EVIDENCE",
"type_str": "figure",
"uris": null,
"num": null
},
"TABREF0": {
"html": null,
"type_str": "table",
"num": null,
"content": "<table><tr><td>[In its future population estimates'] [made</td><td>(3)</td></tr><tr><td>public last year, 2</td><td/></tr><tr><td>(2)</td><td/></tr></table>",
"text": "The Ministry of Health and Welfare last year revealed I ] [population of future estimate according to 2] [in future 1.499 persons as the lowest s] [that after *SAB* rising to turn that 4] [*they* estimated but s ] [already the estimate misses a point ~] [prediction became. 7] ] [the Ministry of Health and Welfare predicted that the SAB would drop to a new low of 1.499 in the future, s) [but would make a comeback after that, 4] [increasing once again, s ] [However, it looks as if that prediction will be quickly shattered. 6]"
},
"TABREF5": {
"html": null,
"type_str": "table",
"num": null,
"content": "<table><tr><td/><td/><td/><td/><td>RtlauoaRcaa\u00a2 e</td></tr><tr><td>~oo</td><td/><td/><td/><td/></tr><tr><td>440o</td><td/><td/><td/><td/></tr><tr><td>~oo</td><td/><td/><td/><td/></tr><tr><td>38 oo</td><td/><td/><td/><td/></tr><tr><td>~oa</td><td/><td/><td/><td/></tr><tr><td>I</td><td>I</td><td>I</td><td>I</td><td>tC~ xlO 3</td></tr></table>",
"text": "Performance of the classifiers"
},
"TABREF6": {
"html": null,
"type_str": "table",
"num": null,
"content": "<table/>",
"text": "Rule examples for the Relation-Reduce classifier."
},
"TABREF8": {
"html": null,
"type_str": "table",
"num": null,
"content": "<table/>",
"text": "Confusion matrix for the Main Action Type classifier."
}
}
}
} |