File size: 71,694 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 |
{
"paper_id": "2019",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T07:30:29.348362Z"
},
"title": "Building Discourse Parser for Thirukkural",
"authors": [
{
"first": "R",
"middle": [],
"last": "Anita",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "SRM Institute of Science and Technology",
"location": {
"addrLine": "Kattankulathur-603 203",
"settlement": "Tamilnadu",
"country": "India"
}
},
"email": ""
},
{
"first": "C",
"middle": [
"N"
],
"last": "Subalalitha",
"suffix": "",
"affiliation": {
"laboratory": "",
"institution": "SRM Institute of Science and Technology",
"location": {
"addrLine": "Kattankulathur-603 203",
"settlement": "Tamilnadu",
"country": "India"
}
},
"email": "subalalitha@gmail.com"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "Thirukkural is one of the famous Tamil Literatures in the world. It was written by Thiruvalluvar, and focuses on ethics and morality. It provides all possible solutions to lead a successful and a peaceful life fitting any generation. It has been translated into 82 global languages, which necessitate the access of Thirukkural in any language on the World Wide Web (WWW) and processing the Thirukkural computationally. This paper aims at constructing the Thirukkural Discourse Parser which finds the semantic relations in the Thirukkurals which can extract the hidden meaning in it and help in utilizing the same in various Natural Language Processing (NLP) applications, such as, Summary Generation Systems, Information Retrieval (IR) Systems and Question Answering (QA) Systems. Rhetorical Structure Theory (RST) is one of the discourse theories, which is used in NLP to find the coherence between texts. This paper finds the relation within the Thriukkurals and the discourse structure is created using the Thirukkural Discourse Parser. The resultant discourse structure of Thirukkural can be indexed and further be used by Summary Generation Systems, IR Systems and QA Systems. This facilitates the end user to access Thirukkural on WWW and get benefited. This Thirukkural Discourse Parser has been tested with all 1330 Thirukurals using precision and recall.",
"pdf_parse": {
"paper_id": "2019",
"_pdf_hash": "",
"abstract": [
{
"text": "Thirukkural is one of the famous Tamil Literatures in the world. It was written by Thiruvalluvar, and focuses on ethics and morality. It provides all possible solutions to lead a successful and a peaceful life fitting any generation. It has been translated into 82 global languages, which necessitate the access of Thirukkural in any language on the World Wide Web (WWW) and processing the Thirukkural computationally. This paper aims at constructing the Thirukkural Discourse Parser which finds the semantic relations in the Thirukkurals which can extract the hidden meaning in it and help in utilizing the same in various Natural Language Processing (NLP) applications, such as, Summary Generation Systems, Information Retrieval (IR) Systems and Question Answering (QA) Systems. Rhetorical Structure Theory (RST) is one of the discourse theories, which is used in NLP to find the coherence between texts. This paper finds the relation within the Thriukkurals and the discourse structure is created using the Thirukkural Discourse Parser. The resultant discourse structure of Thirukkural can be indexed and further be used by Summary Generation Systems, IR Systems and QA Systems. This facilitates the end user to access Thirukkural on WWW and get benefited. This Thirukkural Discourse Parser has been tested with all 1330 Thirukurals using precision and recall.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "Tamil literature has so many nuggets hidden in it which need to be explored for the goodness of the society. One of the ways to explore the Tamil literature is to make it easily accessible on the World Wide Web (WWW). For instance, Thirukkural is one of the famous Tamil literatures in the world and it is respected by people across the globe. In order to make it to reach to all people, it should be made available on the web. This makes necessary to process it computationally. Hence, this paper proposes a methodology to perform a discourse analysis of Thirukkural, which aids in exploring its semantics and also organizing it on the web.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Natural Language Processing (NLP) is the process of interaction between computer and human or natural languages. Analysis of text can be done at various levels namely, word, clause, sentence, paragraph and document. Discourse analysis is used for analyzing the text beyond the clause level. The proposed work attempts to extract the relations found within the Thirukkural.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Discourse structure of a text can be built by using a popular theory called, Rhetorical Structure Theory (RST) (Thompson and Mann, 1987; Mann and Thompson, 1988) . Using RSTbased discourse relations, the RST captures the coherence among the Natural Language (NL) text spans. The coherence can be found between two or more text fragments. The text fragments could be within a sentence, across sentences, across paragraphs and even across documents.",
"cite_spans": [
{
"start": 111,
"end": 136,
"text": "(Thompson and Mann, 1987;",
"ref_id": null
},
{
"start": 137,
"end": 161,
"text": "Mann and Thompson, 1988)",
"ref_id": "BIBREF8"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "In this paper, each Thirukkural is considered as a sentence and discourse parser is built using RST. The contributions of this paper are twofold. 1) Finding feature set using rule based approach. 2) Building Discourse parser by identifying discourse relations.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The rest of the paper is organized as follows. Section 2 describes background details. Section 3 describes the related work. Section 4 discusses the proposed work. Section 5 gives details on the evaluation of the proposed technique. Section 6 presents the conclusion and future works.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "This section describes about the Thirukkural and the basics of RST based Discourse Parsing.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Background",
"sec_num": "2"
},
{
"text": "Thirukkural consists of 1330 couplets or Thirukkurals. They are classified into three sections and 133 chapters. Each chapter in Thirukkural has a specific subject and consists of ten couplets or Thirukkurals. A couplet consists of two lines. Each Thirukkural or couplet is formed with seven cirs (words). First line of the couplet consists of four cirs and the second line of the couplet consists of three cirs. A single Tamil word or a combination of two or more Tamil words forms a cir.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Thirukkural",
"sec_num": "2.1."
},
{
"text": "RST is a descriptive theory which focuses on the organization of the natural language. It was proposed by Bill Mann, Sandy Thompson, and Christian Matthiessen at the University of Southern California (Thompson and Mann, 1987; Mann and Thompson, 1988) . It identifies the coherence between the text spans using discourse relations and forms a discourse structure called rhetorical structure. The discourse units are Nucleus, Satellite and Discourse Relations. The nucleus carries the necessary information and the satellite carries the additional information supporting the nucleus.",
"cite_spans": [
{
"start": 106,
"end": 136,
"text": "Bill Mann, Sandy Thompson, and",
"ref_id": null
},
{
"start": 200,
"end": 225,
"text": "(Thompson and Mann, 1987;",
"ref_id": null
},
{
"start": 226,
"end": 250,
"text": "Mann and Thompson, 1988)",
"ref_id": "BIBREF8"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Rhetorical Structure Theory",
"sec_num": "2.2."
},
{
"text": "Discourse relations are organized into three categories, namely, subject matter, presentational, and multinuclear. In subject matter relations, satellite is a request or problem posed by the reader, i.e. satisfied or solved by nucleus. Elaboration, evaluation and condition are some of the subject matter relations. In presentational relations, satellite increases reader's inclination in accepting the facts stated in nucleus. Antithesis, background and enablement are some of the presentational relations. In multinuclear relations, two nuclei are connected instead of one nucleus and one satellite. Conjunction, contrast and sequence are some of the multinuclear relations. Figure 1 shows an example of how the nucleus, satellite, and the discourse relation are identified for an English sentence in Example 1.",
"cite_spans": [],
"ref_spans": [
{
"start": 677,
"end": 685,
"text": "Figure 1",
"ref_id": "FIGREF0"
}
],
"eq_spans": [],
"section": "Rhetorical Structure Theory",
"sec_num": "2.2."
},
{
"text": "Example 1 Raj sings well but he could not win the contest.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Rhetorical Structure Theory",
"sec_num": "2.2."
},
{
"text": "Raj sings well Satellite:",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Nucleus:",
"sec_num": null
},
{
"text": "he could not win the contest Discourse Relation: Antithesis In Example 1, the sentence holds antithesis relation. It is identified by the signal word but. \"Raj sings well\" is the nucleus, because it represents the ideas favored by the author. \"He could not win the contest\" is the satellite, because it represents the ideas disfavored by the author. These NRS sequences capture the inherent semantics in the texts which is applied to the Thirukkural couplets by the proposed approach.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Nucleus:",
"sec_num": null
},
{
"text": "Subba and Di 2009found discourse relations by using shift reduce parsing model and WordNet. The linguistic cues were used as features. The document was analyzed at sentence level. Hernault et al. (2010) constructed discourse parser by building discourse tree using Support Vector Machine Classifier. The document was analyzed beyond the sentence level and the combination of syntactic and lexical features such as words, POS tags and lexical heads were used as feature sets. Hernault et al. (2010a) used a semi-supervised method called Feature Vector Extension for discourse relation classification. The method was based on the analysis of co-occurring features present in unlabelled data, which was then taken into account for extending the feature vectors given to a classifier. The word pairs, production rules from parse trees and Lexico-Syntactic context at the border between two units of text were used as features for the algorithm. Sucheta et al. (2011) identified explicit discourse connectives for Penn Discourse Tree Bank (PDTB). They proposed shallow discourse parsing for performing token level argument segmentation. The document was analyzed at sentence level. The lexical, syntactic and semantic features were used as features. Sucheta et al. (2012) improved a shallow discourse parser by using a constraint-based method based on conditional random fields and the recall was improved. Sucheta, Giuseppe, and Richard (2012) constructed a parser which uses local constraints and then global constraints. They analyzed the text at the inter sentence level and they used the lexico-syntactic features. 2013used Tamil and Sanskrit literature concepts called suthras and sangatis, along with the current-day text processing theories namely, RST, Universal Networking Language for identifying semantic indices for Tamil documents. Suthras are used for representing the text in a crisp manner.",
"cite_spans": [
{
"start": 180,
"end": 202,
"text": "Hernault et al. (2010)",
"ref_id": "BIBREF5"
},
{
"start": 475,
"end": 498,
"text": "Hernault et al. (2010a)",
"ref_id": "BIBREF4"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "3"
},
{
"text": "Sobha et al. 2014and Sobha and Patnaik (2004) proposed automatic identification of connectives and their arguments for the Indian languages Hindi, Malayalam and Tamil. They used Conditional Random Fields machine learning technique. They used 3000 sentences from a health domain as a corpus. Sobha et al. 2014, annotated the three language corpus, namely Tamil Hindi and Malayalam, with the discourse relations. Lin et al. (2014) constructed an end-to-end discourse parser in the PDTB style. Their parser identified all discourse and non-discourse relations, labeled the arguments, and found the sense of relation between arguments. The document was analyzed at paragraph level. The lexical, syntactic and semantic features were used as features.",
"cite_spans": [
{
"start": 21,
"end": 45,
"text": "Sobha and Patnaik (2004)",
"ref_id": "BIBREF11"
},
{
"start": 411,
"end": 428,
"text": "Lin et al. (2014)",
"ref_id": "BIBREF6"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Subalalitha and Ranjani Parthasarathi",
"sec_num": null
},
{
"text": "Yangfeng and Jacob (2014) transformed surface features into a latent space by using a representation learning approach that facilitates RST discourse parsing. They used shift reduced discourse parser and analyzed the document at sentence level.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Subalalitha and Ranjani Parthasarathi",
"sec_num": null
},
{
"text": "Uladzimir et al. (2015) segmented the German text for the RST-based discourse parsing. They analyzed the text at sentence level. Parminder et al. (2015) proposed document level sentiment analysis using RST discourse parsing and recursive neural network. They analyzed the text at document level and lexical features were used as features. Subalalitha and Ranjani Parthasarathi (2015) found 13 RST Relations in Tamil documents. The Na\u00efve Bayes probabilistic classifier machine learning algorithm was used and the Tamil documents were analyzed beyond the sentence level. The high level semantic features were used by their discourse parser, which were inherited from UNL to construct rhetorical structure trees.",
"cite_spans": [
{
"start": 339,
"end": 383,
"text": "Subalalitha and Ranjani Parthasarathi (2015)",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Subalalitha and Ranjani Parthasarathi",
"sec_num": null
},
{
"text": "Manfred et al. (2016) annotated the corpus with two theories, namely, RST and Segmented Discourse Representation Theory. It was also annotated with the argumentation annotation. The document was analyzed at sentence level. The syntactic and semantic features were used as feature set.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Subalalitha and Ranjani Parthasarathi",
"sec_num": null
},
{
"text": "Yangfeng et al. 2016proposed a latent variable recurrent neural network for finding the discourse relation between adjacent sentences. They analyzed the text at inter sentence level and they have used lexical features. Yangfeng and Noah (2017) proposed text categorization by using recursive neural network and RST. The document was analyzed at sentence level.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Subalalitha and Ranjani Parthasarathi",
"sec_num": null
},
{
"text": "It can be observed that, the existing discourse methodologies analyzed the text in English documents and expository type Tamil documents. This paper proposes a discourse methodology that makes use of RST to identify the semantic relations/discourse relations from a Tamil literature text which lacks a regular pattern for semantic analysis. Unlike English which has a fixed SVO (Subject Verb Object) sentence pattern, Tamil expository texts have either SVO or SOV (Subject Object Verb) pattern. Tamil literatures on the other side neither follow SOV nor SVO pattern. Tamil literatures also have a relatively rich set of morphological variants (Anand et al., 2010; Goldsmith, 2001 ). This makes the processing of Tamil literature more complex than processing the expository Tamil documents. This paper focuses on finding discourse relations in a Tamil literary work called, Thirukkural, which has the structure of classic Tamil language poetry form, called venba. Venba style Tamil literature consists of lines between two and twelve. Expository Tamil documents have the cue words in middle of the sentence. It is not difficult to find the nucleus satellite identification for expository type of texts, whereas the cue words in Tamil literature specifically in venba style of texts will be present in any part of the sentence. If the cue word is present in the middle of the Thirukkural couplet, it is not difficult to find the nucleus satellite identification. If it is present at either end of the Thirukkural couplet then it is difficult to find the nucleus satellite identification. The proposed Thirukkural Discourse Parser handles these cases to an extent which is discussed in the upcoming sections.",
"cite_spans": [
{
"start": 643,
"end": 663,
"text": "(Anand et al., 2010;",
"ref_id": "BIBREF1"
},
{
"start": 664,
"end": 679,
"text": "Goldsmith, 2001",
"ref_id": "BIBREF3"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Subalalitha and Ranjani Parthasarathi",
"sec_num": null
},
{
"text": "The Tamil Thirukkural couplets are given as input. Initially the cue phrases or signal words are identified in each Thirukkural. RST based discourse relations are identified by Thirukkural Discourse Parser based on the cue words and semantics. Then the Nucleus and Satellites are identified for each Thirukkural. Finally, the NRS sequences are identified as output from the Thirukkural Discourse Parser.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Proposed Work",
"sec_num": "4"
},
{
"text": "The connectives connecting two clauses of the Thirukkural are used as the feature set as they signal a discourse relation. An analysis of the 600 Thirukkural has been done and the feature set for each discourse relation has been identified. For condition relation, 108 features have been identified; for evidence relation, 36 features have been identified; for contrast relation, 37 features have been identified; for enablement relation, 24 features have been identified; for background relation, 9 features have been identified. The feature sets, cue words and signal words are interchangeably used in this paper. The part of the cue words are appeared in Table 1 . For example, '\u0b87\u0bb2\u0bb5\u0bc7 (Ilave-If not)', '\u0b86\u0baf\u0bbf\u0ba9\u0bcd (Ayin-If)', and '\u0b86\u0bb1\u0bcd \u0bb1\u0bbf\u0ba9\u0bcd (Arrin-If someone did)' are some of the cue words commonly appeared in Thirukkural.",
"cite_spans": [],
"ref_spans": [
{
"start": 658,
"end": 665,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Feature Sets",
"sec_num": "4.1."
},
{
"text": "The discourse relations namely, condition, evidence, contrast, enablement and background are identified by the Thirukkural Discourse Parser. A cue word may either be a single word which can be explicitly identified by the Thirukkural Discourse Parser or it may be a case suffix which may have to be split by the morphological analyzer (Anandan et al., 2001 ).",
"cite_spans": [
{
"start": 335,
"end": 356,
"text": "(Anandan et al., 2001",
"ref_id": "BIBREF0"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Discourse Relation Identification",
"sec_num": "4.2."
},
{
"text": "If the cue words explicitly appear in the Thirukkural, then the RST based discourse relations are identified using the signal words in Table 1 . The cue words are given in Tamil along with their English transliteration and English meaning.",
"cite_spans": [],
"ref_spans": [
{
"start": 135,
"end": 142,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Discourse Relation Identification",
"sec_num": "4.2."
},
{
"text": "If the cue words do not explicitly appear in the Thirukkural, then the morphological analyzer is used for finding the cue words. For example, in the word '\u0b8e\u0bb4\u0bc1\u0ba4\u0bcd \u0ba4\u0ba4\u0bb2\u0bcd \u0bb2\u0bbe\u0bae\u0bcd (Eluttellam-All the letters) ', the cue word '\u0b8e\u0bb2\u0bcd \u0bb2\u0bbe\u0bae\u0bcd (Ellam-Everything)' is a case suffix and so the morphological analyzer is used to isolate the cue word ('\u0b8e\u0bb4\u0bc1\u0ba4\u0bcd \u0ba4\u0bc1+ \u0b8e\u0bb2\u0bcd \u0bb2\u0bbe\u0bae\u0bcd '). Now the cue word '\u0b8e\u0bb2\u0bcd \u0bb2\u0bbe\u0bae\u0bcd (Ellam-Everything)' can be used for identifying the RST based discourse relation. ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Discourse Relation Identification",
"sec_num": "4.2."
},
{
"text": "The cue words are appeared anywhere in the Thirukkural. In most of the Thirukkural, it is appeared in the middle. The text before the cue word is considered as Clause1 and the text after the cue word is considered as Clause2. Clause1 and Clause2 can be indicated as nucleus and satellite.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Nuclearity Identification",
"sec_num": "4.3."
},
{
"text": "In few Thirukkural, Clause1 can act as the nucleus and Clause2 can act as the satellite. And in others, they may be vice versa. Hence, it is identified separately and the NRS sequences are identified accordingly. Table 2 lists some of the cue words appear in Thirukkural in which Clause1 and Clause2 are categorized as nucleus and satellite.",
"cite_spans": [],
"ref_spans": [
{
"start": 213,
"end": 220,
"text": "Table 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Nuclearity Identification",
"sec_num": "4.3."
},
{
"text": "Example 2 in Figure 2 shows NRS sequence identified by the Thirukkural Discourse Parser.",
"cite_spans": [],
"ref_spans": [
{
"start": 13,
"end": 21,
"text": "Figure 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Nuclearity Identification",
"sec_num": "4.3."
},
{
"text": "Clause1-Nucleus, Clause2-Satellite ",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Relation",
"sec_num": null
},
{
"text": "The Thirukkural in Figure 2 is given as the input for the Thirukkural Discourse Parser. This Thirukkural has the cue word '\u0bb5\u0bc6\u0bbe\u0bb2 (Pola-Like)' explicitly. This cue word is used for identifying the Enablement relation. The Thirukkural Discourse Parser identifies 'Enablement' as the discourse relation, '\u0b86\u0b99\u0bcd \u0bb5\u0b95 \u0b87\u0b9f\u0bc1\u0b95\u0bcd \u0b95\u0ba3 \u0bcd \u0b95\u0bb1\u0bb3\u0bc7\u0ba4\u0bbe\u0bae\u0bcd \u0ba8\u0b9f\u0bcd \u0baa\u0bc1' as nucleus, (as it contains an action), and '\u0b89\u0b9f\u0bc1\u0b95\u0bcd \u0bb1\u0b95 \u0b87\u0bb4\u0ba8\u0bcd \u0ba4\u0bc7\u0ba9\u0bcd \u0bb1\u0b95' as satellite, (as it contains the information for performing the action). Similarly NRS sequences for all such cases present in the Thirukkurals are identified by using the Thirukkural Discourse Parser. Figure 3 shows Nucleus, Satellite and Discourse relation for the Example 2. ",
"cite_spans": [],
"ref_spans": [
{
"start": 19,
"end": 27,
"text": "Figure 2",
"ref_id": null
},
{
"start": 607,
"end": 615,
"text": "Figure 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Table 2. Nucleus and Satellite Identification",
"sec_num": null
},
{
"text": "Utukkai ilantavan kaipola anke itukkan kalaivatam natpu.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "English Transliteration:",
"sec_num": null
},
{
"text": "True friendship hastens to the rescue of the afflicted as readily as the hand of one whose garment is loosened.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "The features are extracted from 600 couplets. The 1330 Thirukkural couplets are given as the input for the Thirukkural Discourse Parser. The NRS sequences are identified by the Thirukkural Discourse Parser based on the cue words. This work is evaluated using the parameters, precision and recall. Precision (P), and recall (R) values are calculated using equations 1 ( 2)The Table 3 shows the precision and recall of the discourse parser. The total number of NRS sequences retrieved by the proposed Thirukkural Discourse Parser is denoted as -M, total number of relevant NRS sequences actually present in Thirukkural is denoted as -N, and number of relevant NRS sequences retrieved by the proposed Thirukkural Discourse Parser is denoted as -C. The value of the variables, C and N are calculated using human judgement. About six domain experts have calculated these metric N, and the average has been taken and presented in Table 3 . It can be observed from the table that the total number of NRS sequences relevant to the condition relation is 547 and total number of NRS sequences retrieved from the Thirukkural Discourse Parser is 616. This is because more than one cue words belonging to the Condition relation are appeared in some Thirukkurals.",
"cite_spans": [],
"ref_spans": [
{
"start": 375,
"end": 382,
"text": "Table 3",
"ref_id": null
},
{
"start": 924,
"end": 931,
"text": "Table 3",
"ref_id": null
}
],
"eq_spans": [],
"section": "Evaluation",
"sec_num": "5"
},
{
"text": "In Example 3 shown in Figure 5 , two cue words \"\u0b86\u0bb1\u0bcd\u0bb1\u0bbf\u0ba9\u0bcd (Arrin-If someone did)\" and \"\u0baa\u0bc6\u0bb1\u0bbf\u0ba9\u0bcd (Perin-If received)\" have appeared in the Thirukkural. Both cue words belong to the Condition relation. In order to analyze the significance of each feature, the NRS sequences emerging out of two features present in the same Thirukkural is counted and hence, M is greater than N. Similarly, for Evidence relation also M is greater than N. On the other side, in Contrast, Enablement and Background relations, N is more than M.",
"cite_spans": [],
"ref_spans": [
{
"start": 22,
"end": 30,
"text": "Figure 5",
"ref_id": "FIGREF5"
}
],
"eq_spans": [],
"section": "Evaluation",
"sec_num": "5"
},
{
"text": "\u0ba4\u0bc6\u0bb1\u0bbf\u0bc7\u0bb1\u0bbf\u0ba8\u0bcd \u0ba4\u0bc1 \u0b9a\u0bc0\u0bb0 \u0bcd\u0bb1\u0bae \u0bc6\u0baf\u0b95\u0bcd \u0b95\u0bc1\u0bae\u0bcd \u0b85\u0bb1\u0bbf\u0bc7\u0bb1\u0bbf\u0ba8\u0bcd \u0ba4\u0bc1 \u0b86\u0bb1\u0bcd \u0bb1\u0bbf\u0ba9\u0bcd \u0b85\u0b9f\u0b99\u0bcd \u0b95\u0bc6\u0bcd \u0ba4\u0bc6\u0bb1\u0bbf\u0ba9\u0bcd .",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Example 3:",
"sec_num": null
},
{
"text": "Cerivarintu cirmai payakkum arivarintu aarrin atankap perin.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "English Transliteration:",
"sec_num": null
},
{
"text": "Knowing that self-control is knowledge, if a man should control himself, in the prescribed course, such self-control will bring him distinction among the wise. The correctly retrieved Thirukkurals, C is smaller than N in all the discourse relation, this is due to the inability of the discourse parser to extract the NRS sequences using implicit cue words.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "\u0b89\u0bb3\u0bcd \u0bb3\u0bb1\u0bcd \u0b95 \u0b89\u0bb3\u0bcd \u0bb3\u0bae\u0bcd \u0b9a\u0bbf\u0bb1\u0bc1\u0b95\u0bc1\u0bc7 \u0ba4\u0b95\u0bbe\u0bb3\u0bcd \u0bb3\u0bb1\u0bcd \u0b95 \u0b85\u0bb2\u0bcd \u0bb2\u0bb1\u0bcd \u0b95\u0ba3 \u0bcd \u0b86\u0bb1\u0bcd \u0bb1\u0bb1\u0bc1\u0bc6\u0bcd \u0bc6\u0bbe\u0bb0 \u0bcd \u0ba8\u0b9f\u0bcd \u0baa\u0bc1.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Example 4:",
"sec_num": null
},
{
"text": "Ullarka ullam cirukuva kollarka allarka\u1e47 arraruppar natpu.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "English Transliteration:",
"sec_num": null
},
{
"text": "Do not think of things that discourage your mind, nor contract friendship with those who would forsake you in adversity. In Example 4 shown in Figure 6 , the cue word \"\u0bb5\u0bc6\u0bbe\u0bb2 (Pola-Like)\" is implicit. The Enablement relation identification needs additional semantic analysis which is currently not done by the discourse parser.",
"cite_spans": [],
"ref_spans": [
{
"start": 143,
"end": 151,
"text": "Figure 6",
"ref_id": "FIGREF6"
}
],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "In some Thirukkurals, more than one cue words pointing to different relations have appeared. Therefore more than one NRS sequences are identified by the Thirukkural Discourse Parser for the same Thirukkural.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "In Example 5 shown in Figure 7 , two cue words namely, \"\u0b86\u0bb1\u0bcd \u0bb1\u0bbf\u0ba9\u0bcd (Arrin-If someone did)\" and \"\u0bb5\u0bc6\u0bbe\u0bb2\u0bcd (Pol-Like)\" are present.",
"cite_spans": [],
"ref_spans": [
{
"start": 22,
"end": 30,
"text": "Figure 7",
"ref_id": "FIGREF7"
}
],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "\"\u0b86\u0bb1\u0bcd \u0bb1\u0bbf\u0ba9\u0bcd (Arrin-If someone did)\" is a cue word related to Condition relation and \"\u0bb5\u0bc6\u0bbe\u0bb2\u0bcd (Pol-Like)\" is a cue word related to Evidence relation. So the Condition and Evidence relations are identified by the Thirukkural Discourse Parser.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "\u0b92\u0bb0\u0bc1\u0bb1\u0bae\u0baf\u0bc1\u0bb3\u0bcd \u0b86\u0bb1\u0bae\u0bb5\u0bc6\u0bbe\u0bb2\u0bcd \u0b90\u0ba8\u0bcd \u0ba4\u0b9f\u0b95\u0bcd \u0b95\u0bb2\u0bcd \u0b86\u0bb1\u0bcd \u0bb1\u0bbf\u0ba9\u0bcd \u0b8e\u0bb4\u0bc1\u0ba8\u0bae\u0bcd \u0baf\u0bc1\u0bae\u0bcd \u0b8f\u0bae\u0bbe\u0bc6\u0bcd \u0baa\u0bc1\u0bb1\u0b9f\u0ba4\u0bcd \u0ba4\u0bc1.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Example 5:",
"sec_num": null
},
{
"text": "Orumaiyul amaipol aintatakkal arrin elunamyum emap putaittu.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "English Transliteration:",
"sec_num": null
},
{
"text": "Should one throughout a single birth, like a tortoise keep in his five senses, the fruit of it will prove a safe-guard to him throughout the seven-fold births. The precision and recall values of the discourse parser can further be increased by increasing the feature sets, by incorporating a machine learning algorithm. The efficiency can be increased by finding the discourse relations for the Thirukkurals having implicit cue words. The efficiency of the Thirukkural Discourse Parser also depends on the efficiency of the morphological analyzer. A high level semantic knowledge base such as WordNet (George, 1995) or ontology may improve the efficiency even better.",
"cite_spans": [
{
"start": 601,
"end": 615,
"text": "(George, 1995)",
"ref_id": "BIBREF2"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Meaning in English:",
"sec_num": null
},
{
"text": "Thirukkural has much valuable information that is to be followed by the society. In order to access the Thirukkural on the web, the semantic analysis of the same becomes necessary. This paper makes use of discourse theory, named, RST to do a discourse/semantic analysis on the Thirukkural which will be useful to retrieve the Thirukkural using an Information Retrieval System. Keyword-based Thirukkural search is available but limits the user to retreive the Thirukkural containing only the query words. In this paper, we propose a methodology to construct a discourse parser for Thirukkural which aids in semantic analysis and semantic representation of Thirukkural. This kind of semantic representation can be used for efficient semantic indexing of Thirukkural for better retrieval.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion and Future Works",
"sec_num": "6"
},
{
"text": "Using the results of the proposed discourse parser, an analysis of how Thirukkural is written and organized is also evident which can be useful to write similar works in future. This kind of analysis can also help in automatic author detection of text (Stamatatos, 2008) . This paper focuses on the discourse relations present within a Thirukkural couplet. It may be further extended to find the relations across the Thirukkural couplets. An application that clearly depicts the discourse structure representation is also to be done.",
"cite_spans": [
{
"start": 252,
"end": 270,
"text": "(Stamatatos, 2008)",
"ref_id": "BIBREF14"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion and Future Works",
"sec_num": "6"
}
],
"back_matter": [
{
"text": " Subalalitha, C.N., and Ranjani Parthasarathi. 2015 ",
"cite_spans": [
{
"start": 1,
"end": 51,
"text": "Subalalitha, C.N., and Ranjani Parthasarathi. 2015",
"ref_id": null
}
],
"ref_spans": [],
"eq_spans": [],
"section": "annex",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Morphological Analyzer for Tamil",
"authors": [
{
"first": "P",
"middle": [],
"last": "Anandan",
"suffix": ""
},
{
"first": "Ranjani",
"middle": [],
"last": "Parthasarathy",
"suffix": ""
},
{
"first": "T",
"middle": [
"V"
],
"last": "Geetha",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Anandan, P., Ranjani Parthasarathy, and Geetha, T.V. 2001. Morphological Analyzer for Tamil. ICON 2002, RCILTS-Tamil, Anna University, India.",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "A Sequence Labeling Approach to Morphological Analyzer for Tamil Language",
"authors": [
{
"first": "Anand",
"middle": [],
"last": "Kumar",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Dhanalakshmi",
"suffix": ""
},
{
"first": "V",
"middle": [],
"last": "Soman",
"suffix": ""
},
{
"first": "K",
"middle": [
"P"
],
"last": "Rajendran",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "",
"suffix": ""
}
],
"year": 2010,
"venue": "International Journal on Computer Science and Engineering (IJCSE)",
"volume": "02",
"issue": "06",
"pages": "1944--195",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Anand Kumar, M., Dhanalakshmi, V., Soman, K.P., and Rajendran, S. 2010. A Sequence Labeling Approach to Morphological Analyzer for Tamil Language. International Journal on Computer Science and Engineering (IJCSE), Vol. 02, No. 06, 1944-195.",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "WordNet: A Lexical Database for English",
"authors": [
{
"first": "A",
"middle": [],
"last": "George",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Miller",
"suffix": ""
}
],
"year": 1995,
"venue": "Communications of the ACM",
"volume": "38",
"issue": "11",
"pages": "39--41",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "George, A. Miller. 1995. WordNet: A Lexical Database for English. Communications of the ACM Vol. 38, No. 11:39-41.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Unsupervised learning of the morphology of a natural language",
"authors": [
{
"first": "J",
"middle": [],
"last": "Goldsmith",
"suffix": ""
}
],
"year": 2001,
"venue": "Computational Linguistics",
"volume": "27",
"issue": "",
"pages": "153--198",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Goldsmith, J. 2001. Unsupervised learning of the morphology of a natural language. Computational Linguistics, 27, 153-198.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "A semi-supervised approach to improve classification of infrequent discourse relations using feature vector extension",
"authors": [
{
"first": "H",
"middle": [],
"last": "Hernault",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "Bollegala",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Ishizuka",
"suffix": ""
}
],
"year": 2010,
"venue": "Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing",
"volume": "",
"issue": "",
"pages": "399--409",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hernault, H., Bollegala, D., and Ishizuka, M. 2010a. A semi-supervised approach to improve classification of infrequent discourse relations using feature vector extension. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics: Cambridge, MA, pp. 399-409.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "HILDA: A discourse parser using support vector machine classification",
"authors": [
{
"first": "H",
"middle": [],
"last": "Hernault",
"suffix": ""
},
{
"first": "H",
"middle": [],
"last": "Predinger",
"suffix": ""
},
{
"first": "D",
"middle": [
"A"
],
"last": "Duverle",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Ishizuka",
"suffix": ""
}
],
"year": 2010,
"venue": "Dialogue and Discourse",
"volume": "1",
"issue": "3",
"pages": "1--33",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hernault, H., Predinger, H., Duverle, D.A., and Ishizuka, M. 2010. HILDA: A discourse parser using support vector machine classification. Dialogue and Discourse, vol.1, no.3, pp. 1-33.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "A PDTB-styled end-to-end discourse parser",
"authors": [
{
"first": "Z",
"middle": [],
"last": "Lin",
"suffix": ""
},
{
"first": "H",
"middle": [],
"last": "Ng",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Kan",
"suffix": ""
}
],
"year": 2014,
"venue": "Natural Language Engineering",
"volume": "20",
"issue": "2",
"pages": "151--184",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Lin, Z., NG, H., and Kan, M. 2014. A PDTB-styled end-to-end discourse parser. Natural Language Engineering, 20(2), 151-184.",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Parallel discourse annotations on a corpus of short texts",
"authors": [
{
"first": "Manfred",
"middle": [],
"last": "Stede",
"suffix": ""
},
{
"first": "Stergos",
"middle": [],
"last": "Afantenos",
"suffix": ""
},
{
"first": "Andreas",
"middle": [],
"last": "Peldszus",
"suffix": ""
},
{
"first": "Nicholas",
"middle": [],
"last": "Asher",
"suffix": ""
},
{
"first": "Jeremy",
"middle": [],
"last": "Perret",
"suffix": ""
}
],
"year": 2016,
"venue": "Proceedings of the International Conference on Language Resources and Evaluation (LREC)",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Manfred Stede, Stergos Afantenos, Andreas Peldszus,Nicholas Asher, and Jeremy Perret. 2016. Parallel discourse annotations on a corpus of short texts. In Proceedings of the International Conference on Language Resources and Evaluation (LREC), Portoroz.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Rhetorical structure theory: Toward a functional theory of text organization",
"authors": [
{
"first": "W",
"middle": [
"C"
],
"last": "Mann",
"suffix": ""
},
{
"first": "S",
"middle": [
"A"
],
"last": "Thompson",
"suffix": ""
}
],
"year": 1988,
"venue": "Text-interdisciplinary Journal for the Study of Discourse",
"volume": "8",
"issue": "3",
"pages": "243--281",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Mann, W.C. and Thompson, S.A., 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text-interdisciplinary Journal for the Study of Discourse, 8(3), pp.243-281.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Better document-level sentiment analysis from RST discourse parsing",
"authors": [
{
"first": "Parminder",
"middle": [],
"last": "Bhatia",
"suffix": ""
},
{
"first": "Yangfeng",
"middle": [],
"last": "Ji",
"suffix": ""
},
{
"first": "Jacob",
"middle": [],
"last": "Eisenstein",
"suffix": ""
}
],
"year": 2015,
"venue": "EMNLP",
"volume": "",
"issue": "",
"pages": "2212--2218",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Parminder Bhatia, Yangfeng Ji, and Jacob Eisenstein. 2015. Better document-level sentiment analysis from RST discourse parsing. In EMNLP, pages 2212-2218, 2015.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "Sentential structure and discourse parsing",
"authors": [
{
"first": "L",
"middle": [
"C"
],
"last": "Polanyi",
"suffix": ""
},
{
"first": "M",
"middle": [
"H"
],
"last": "Culy",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Van Den",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Berg",
"suffix": ""
},
{
"first": "G",
"middle": [
"L"
],
"last": "Thione",
"suffix": ""
},
{
"first": "Ahn",
"middle": [],
"last": "",
"suffix": ""
},
{
"first": "D",
"middle": [],
"last": "",
"suffix": ""
}
],
"year": 2004,
"venue": "Proceedings of the ACL 2004 Workshop on Discourse Annotation",
"volume": "",
"issue": "",
"pages": "80--87",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Polanyi, L.C., Culy, M.H., Van Den Berg, Thione, G.L., and Ahn, D. 2004. Sentential structure and discourse parsing. In Proceedings of the ACL 2004 Workshop on Discourse Annotation, Barcelona, Spain, pp. 80-87.",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "Discourse Connectives and Their Arguments in Malayalam",
"authors": [
{
"first": "Devi",
"middle": [],
"last": "Sobha Lalitha",
"suffix": ""
},
{
"first": "B",
"middle": [
"N"
],
"last": "Patnaik",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sobha Lalitha Devi and B. N. Patnaik (2004), Discourse Connectives and Their Arguments in Malayalam\", 24th South Asian Language Analysis, University of Stony Brook , New York, November 19-21",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Discourse Tagging for Indian Languages",
"authors": [
{
"first": "Lakshmi",
"middle": [],
"last": "Sobha Lalitha Devi",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "",
"suffix": ""
},
{
"first": "Sindhuja",
"middle": [],
"last": "Gopalan",
"suffix": ""
}
],
"year": 2014,
"venue": "Computational Linguistics and Intelligent Text Processing",
"volume": "8403",
"issue": "",
"pages": "469--480",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sobha Lalitha Devi, Lakshmi S and Sindhuja Gopalan(2014). \"Discourse Tagging for Indian Languages\", In A. Gelbukh (ed), Computational Linguistics and Intelligent Text Processing, Springer LNCS Vol 8403, pp. 469-480.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Automatic Identification of Discourse Relations in Indian Languages",
"authors": [
{
"first": "Sindhuja",
"middle": [],
"last": "Sobha Lalitha Devi",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Gopalan",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Lakshmi",
"suffix": ""
}
],
"year": 2014,
"venue": "proceedings of 2nd Workshop on Indian Language Data: Resources and Evaluation",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Sobha Lalitha Devi, Sindhuja Gopalan, Lakshmi S (2014). \"Automatic Identification of Discourse Relations in Indian Languages\", In proceedings of 2nd Workshop on Indian Language Data: Resources and Evaluation, Organized under LREC2014, Reykjavik, Iceland",
"links": null
},
"BIBREF14": {
"ref_id": "b14",
"title": "Author identification: Using text sampling to handle the class imbalance problem",
"authors": [
{
"first": "E",
"middle": [],
"last": "Stamatatos",
"suffix": ""
}
],
"year": 2008,
"venue": "Information Processing and Management",
"volume": "44",
"issue": "2",
"pages": "790--799",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Stamatatos, E. 2008. Author identification: Using text sampling to handle the class imbalance problem. Information Processing and Management, 44(2), 790-799.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "A Unique Indexing Technique for Indexing Discourse Structures",
"authors": [
{
"first": "C",
"middle": [
"N"
],
"last": "Subalalitha",
"suffix": ""
},
{
"first": "Ranjani",
"middle": [],
"last": "Parthasarathi",
"suffix": ""
}
],
"year": 2013,
"venue": "Journal of Intelligent Systems",
"volume": "23",
"issue": "3",
"pages": "231--243",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Subalalitha, C.N., and Ranjani Parthasarathi 2013. A Unique Indexing Technique for Indexing Discourse Structures. Journal of Intelligent Systems. Volume 23, Issue 3, Pages 231-243.",
"links": null
}
},
"ref_entries": {
"FIGREF0": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "NRS Sequence for Example 1."
},
"FIGREF1": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Find cue words in all Thirukkural Couplets (ii) Store it separately corresponding to the relation (iii) for Thirukkural couplets = 1 to 1330 do if cue word is present in the Thirukkural then identify the Discourse relation display the NRS sequences end if cue word is not present in the Thirukkural then use morphological analyzer to find the cue word identify the Discourse relation display the NRS sequences end end"
},
"FIGREF2": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Algorithm for Thirukkural Discourse Parser"
},
"FIGREF3": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Number of relevant NRS sequences retrieved, (C)Total number of relevant NRS sequences present,(N)"
},
"FIGREF5": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Example 3."
},
"FIGREF6": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Example 4."
},
"FIGREF7": {
"type_str": "figure",
"num": null,
"uris": null,
"text": "Example 5."
},
"TABREF0": {
"html": null,
"content": "<table><tr><td>S.</td><td/><td/></tr><tr><td>No.</td><td>Relation</td><td>List of Cue words</td></tr><tr><td>1</td><td>Condition</td><td>\u0b87\u0bb2\u0bb5\u0bc7 (Ilave-If not),</td></tr><tr><td/><td/><td>\u0b8e\u0ba9\u0bcd \u0ba9\u0bc1\u0bae\u0bcd (Ennum-The),</td></tr><tr><td/><td/><td>\u0ba4\u0bc6\u0bb1\u0bbf\u0ba9\u0bcd (Perin-If received),</td></tr><tr><td/><td/><td>\u0b8e\u0ba9\u0bcd \u0ba9\u0bbe\u0bae\u0bcd (Ennam-If),</td></tr><tr><td/><td/><td>\u0b86\u0baf\u0bbf\u0ba9\u0bcd (Ayin-If),</td></tr><tr><td/><td/><td>\u0b86\u0bb1\u0bcd \u0bb1\u0bbf\u0ba9\u0bcd (Arrin-If someone did)</td></tr><tr><td>2</td><td>Evidence</td><td>\u0bb5\u0bc7\u0ba3 \u0bcd \u0b9f\u0bbe (Venta-Do Not),</td></tr><tr><td/><td/><td>\u0b85\u0ba4\u0bc1\u0bc7\u0bb2\u0bcd \u0bb2\u0ba4\u0bc1</td></tr><tr><td/><td/><td>(Atuvallatu -That is not),</td></tr><tr><td/><td/><td>\u0ba4\u0bbe\u0ba9\u0bcd (Tan-Just),</td></tr><tr><td/><td/><td>\u0bb5\u0bc6\u0bbe\u0ba9\u0bcd \u0bb1\u0bc1 (Ponru-Like ),</td></tr><tr><td/><td/><td>\u0bb5\u0ba4\u0bb0\u0bbf\u0ba9\u0bcd (Terin-Selection ),</td></tr><tr><td/><td/><td>\u0b8e\u0b99\u0bcd \u0b99\u0ba9\u0bae\u0bcd (Ennanam-How)</td></tr><tr><td>3</td><td>Contrast</td><td>\u0b85\u0bb0\u0bbf\u0baf (Ariya -Rare),</td></tr><tr><td/><td/><td>\u0bae\u0bb1\u0bcd \u0ba4\u0bb1\u0bb2\u0bcd \u0bb2\u0bbe\u0bae\u0bcd</td></tr><tr><td/><td/><td>(Marrellam-On every),</td></tr><tr><td/><td/><td>\u0b86\u0ba4\u0bb2\u0bcd (Atal-Therefore),</td></tr><tr><td/><td/><td>\u0b89\u0baf\u0bcd \u0b95\u0bcd \u0b95\u0bc1\u0bae\u0bcd (Uykkum-That derived)</td></tr><tr><td>4</td><td>Enablement</td><td>\u0b8e\u0bb2\u0bcd \u0bb2\u0bbe\u0bae\u0bcd (Ellam-Everything),</td></tr><tr><td/><td/><td>\u0b85\u0bc7\u0bb0\u0bc1\u0bb3\u0bcd \u0bb3\u0bc1\u0bae\u0bcd (Avarullum-Plunge),</td></tr><tr><td/><td/><td>\u0bae\u0ba9\u0bcd \u0bb1 (Manra-House),</td></tr><tr><td/><td/><td>\u0bb5\u0bc6\u0bbe\u0bb2(Pola-Like)</td></tr><tr><td>5</td><td>Background</td><td>\u0b8e\u0ba9\u0bbf\u0ba9\u0bc1\u0bae\u0bcd (Eninum-However),</td></tr><tr><td/><td/><td>\u0ba4\u0bc6\u0bb2\u0bcd \u0bb2\u0bbe\u0ba4\u0bc1 (Cellatu-Invalid),</td></tr><tr><td/><td/><td>\u0b87\u0ba9\u0bbf\u0bb5\u0ba4 (Inite-Greeter),</td></tr><tr><td/><td/><td>\u0b85\u0bb1\u0bcd \u0bb1\u0bc1 (Arru-Without),</td></tr><tr><td/><td/><td>\u0b89\u0bb1\u0bb1\u0baf\u0bc1\u0bae\u0bcd (Uraiyum-Freezing)</td></tr></table>",
"num": null,
"type_str": "table",
"text": ""
},
"TABREF2": {
"html": null,
"content": "<table><tr><td>Example 2:</td></tr></table>",
"num": null,
"type_str": "table",
"text": "\u0b89\u0b9f\u0bc1\u0b95\u0bcd \u0bb1\u0b95 \u0b87\u0bb4\u0ba8\u0bcd \u0ba4\u0bc7\u0ba9\u0bcd \u0bb1\u0b95\u0bb5\u0bc6\u0bbe\u0bb2 \u0b86\u0b99\u0bcd \u0bb5\u0b95 \u0b87\u0b9f\u0bc1\u0b95\u0bcd \u0b95\u0ba3 \u0bcd \u0b95\u0bb1\u0bb3\u0bc7\u0ba4\u0bbe\u0bae\u0bcd \u0ba8\u0b9f\u0bcd \u0baa\u0bc1."
},
"TABREF3": {
"html": null,
"content": "<table><tr><td>Condition</td><td>547 616 514 83.44</td><td>93.97</td></tr><tr><td>Evidence</td><td>248 283 225 79.51</td><td>90.73</td></tr><tr><td>Contrast</td><td>262 216 182 84.26</td><td>75.21</td></tr><tr><td colspan=\"2\">Enablement 189 158 129 81.65</td><td>76.33</td></tr><tr><td colspan=\"2\">Background 176 145 114 78.62</td><td>73.08</td></tr></table>",
"num": null,
"type_str": "table",
"text": ""
}
}
}
} |