Recent Advances in Natural Language Processing: Selected Papers from RANLP'95

RECENT ADVANCES IN NATURAL LANGUAGE PROCESSING......Page 2
Editorial page......Page 3
Title page......Page 4
Copyright page......Page 5
Table of contents......Page 6
Editors\' Foreword......Page 10
I. MORPHOLOGY AND SYNTAX......Page 14
1. Lexicalisation......Page 16
2. Lexicalised Tree-Adjoining Grammar......Page 17
4. Synchronous TAGs......Page 19
5. Viewing lexicalised trees as super parts-of-speech......Page 20
6. LTAGs and explanation-based learning techniques......Page 22
6.1 Implications of LTAG representation for EBL......Page 23
7. LTAGs and Categorial Grammars......Page 24
REFERENCES......Page 26
1. Background......Page 28
2. Case and order in English......Page 30
Subject affinity rules......Page 31
Linear precedence rules......Page 32
3. Case and word order in German......Page 36
4. Implementation......Page 41
5. English is German......Page 44
REFERENCES......Page 45
1. Introduction......Page 48
2. STSGs: Definitions, terminology and properties......Page 49
3. Disambiguating an input sentence......Page 51
4. Experimental results......Page 56
REFERENCES......Page 58
1. Introduction......Page 60
2.1 The overt characteristics of self-repair......Page 61
2.2 The structural characteristics of self-repair......Page 62
3.1 Syntactic well-formedness......Page 63
3.2 Cascaded repair......Page 64
3.4 Parsing self-repairs......Page 65
4. Representing self-repair......Page 66
5.2 Augmentations to the standard algorithm......Page 67
5.3 The augmented algorithm......Page 68
REFERENCES......Page 70
2. Controlled language and grammar checking......Page 72
3.1 Positive and negative detection......Page 73
3.4 Current methods......Page 74
4. Chart parsing with finite state automata......Page 76
5.2 Insertion......Page 78
6. Complexity......Page 80
8. Conclusions......Page 81
REFERENCES......Page 82
1. Model of Applicative and Cognitive Grammar......Page 84
1.1 Categorial grammars......Page 85
1.2 Applicative and Combinatory Categorial Grammar......Page 86
2. Structural reorganisation......Page 89
3. Coordination......Page 91
4. Meta-rules......Page 93
5. Examples......Page 94
REFERENCES......Page 96
1. Introduction......Page 98
2. Ontological engineering for ellipsis resolution......Page 100
3. Functional centering principles......Page 102
4. Grammatical predicates for textual ellipsis......Page 103
5. Text cohesion parsing: Ellipsis resolution......Page 105
7. Conclusion......Page 107
REFERENCES......Page 108
1. Introduction......Page 110
2. Lexical transducers......Page 112
3. The standard analyser......Page 113
4. The analysis and correction of linguistic variants......Page 115
5. The analysis of unknown words......Page 116
6. Conclusions......Page 119
REFERENCES......Page 120
II. SEMANTICS AND DISAMBIGUATION......Page 122
1. Introduction......Page 124
2. Vector-representation of word meaning......Page 125
2.1 From an English dictionary to P-vectors......Page 126
2.2 From P-vectors to Q-vectors......Page 127
3. Adaptive scaling of the semantic space......Page 128
3.1 Semantic subspaces......Page 129
3.2 Adaptive scaling......Page 130
4. Examples of measuring the word distance......Page 131
5. Evaluation through word prediction......Page 133
6.1 Semantic vectors......Page 134
7. Conclusion......Page 135
REFERENCES......Page 136
1. Introduction......Page 138
2. Epsilon [Є]: Knowledge acquisition as an evolutionary process......Page 139
3.1 POS information......Page 141
4.1 Framework......Page 142
5.1 Inverse KWIC......Page 144
6. Central knowledge base......Page 145
7.2 Dynamic context matching......Page 146
REFERENCES......Page 148
1. Sense disambiguation and sense tuning......Page 150
2. A context-based classifier......Page 151
3. Discussion of the results......Page 156
REFERENCES......Page 160
1. Introduction......Page 162
2.1 Basic process......Page 163
2.2 Reduction of ambiguity in concept hypotheses......Page 164
3.1 Basic process......Page 165
3.2 Speech understanding experiments......Page 166
4.1 Basic process......Page 168
4.2 Speech understanding experiments......Page 170
REFERENCES......Page 172
1. Introduction......Page 174
2. WordNet and the semantic concordance......Page 175
3. Conceptual density and word sense disambiguation......Page 176
4. The disambiguation algorithm using conceptual density......Page 179
5. The experiment......Page 181
6. Conclusions......Page 183
REFERENCES......Page 184
1. Introduction......Page 186
2.1 Text representation......Page 187
2.2 The episodic memory......Page 189
3.1 Similarity of TUs......Page 192
3.2 Similarity of slots and similarity of graphs......Page 193
3.3 Memorisation of an episode: The aggregation process......Page 195
4. Conclusion......Page 196
REFERENCES......Page 197
1. Introduction......Page 198
2.1.1 Three levels of granularity for ambiguity labelling......Page 200
2.1.2 Task-derived limitations on utterance-level ambiguities......Page 201
2.1.3 Necessity to consider utterance-level ambiguities in the context offull utterances......Page 202
2.2.2 Computable representations and \'reasonable\' analysers......Page 203
2.2.3 Expectations for a system of manual labelling......Page 204
2.3.1 Proper representations......Page 205
2.4.1 Informal presentation......Page 206
2.4.2 Scope of an ambiguity......Page 207
2.4.3 Occurrence and kernel of an ambiguity......Page 209
3. Attributes and values used in manual labelling......Page 210
3.2.1 Structure of the list and associated separators......Page 211
3.3.2 Headers of ambiguity kernels......Page 212
3.3.3 Obligatory labels......Page 213
3.3.4 Other labels......Page 214
REFERENCES......Page 215
III. DISCOURSE......Page 224
2. Centering model for English analysis......Page 226
2.1 Extension to the centering algorithm......Page 227
3. Local discourse mechanisms in translation......Page 229
4. Ordering of Polish constituents......Page 230
4.1 Ordering criteria......Page 231
4.2 Building on orders of constituents......Page 232
5. Conclusion......Page 233
REFERENCES......Page 235
1. Introduction......Page 238
2. An integrated anaphor resolution approach......Page 239
3. An uncertainty reasoning approach......Page 240
4. The two-engine strategy......Page 242
5. Illustration......Page 243
6. Conclusion......Page 245
REFERENCES......Page 246
1. Introduction......Page 248
2. Topic recognition model......Page 249
3. Text representation......Page 251
4.1 Test setting......Page 253
4.2 Result and analysis......Page 255
5. Conclusion......Page 257
REFERENCES......Page 258
1. Introduction......Page 260
2. Text type, subject matter and theme selection......Page 262
3. Theme selection as interstratal constraints......Page 267
4. Conclusions......Page 269
REFERENCES......Page 270
1. Introduction......Page 272
2. The communication of information......Page 273
3. Rhetorical structure of turns......Page 276
4. An example......Page 278
5. Discerning relevant information......Page 279
6. Conclusions......Page 281
REFERENCES......Page 282
IV. GENERATION......Page 284
1. Introduction......Page 286
2. Generation from non-hierarchical representations......Page 287
3. D-Tree Grammars......Page 290
4. Knowledge sources......Page 292
4.1 Mapping rules......Page 293
5. Sentence generation......Page 294
5.2 Covering the remaining semantics......Page 295
5.3 Completing a derivation......Page 296
6. Example......Page 297
7. Matching the applicability semantics of mapping rules......Page 298
8. Preference-based chart generation......Page 299
10. Discussion......Page 302
11. Conclusion......Page 304
REFERENCES......Page 305
1. Introduction......Page 308
2. LR compilation for parsing......Page 310
3. The semantic head-driven generation algorithm......Page 312
4. Grammar inversion......Page 314
5. LR compilation for generation......Page 316
7. Optimising the generation tables......Page 319
8. An example-based optimisation technique......Page 322
9. Discussion......Page 325
REFERENCES......Page 327
1. Introduction: The speaker\'s problem......Page 330
2. What kind of evidence can we provide in favour of patternmatching?......Page 334
3. Why cognitive linguistics, or, why study natural language inthe realm of cognitive science?......Page 338
4. Where do linguistic structures come from?......Page 340
5. Conceptual structures and syntactic structures are to a greatextent parallel......Page 341
5.1 Discussion......Page 343
6. What do syntactic structures depend upon?......Page 346
7. Prototypical patterns......Page 349
8. Where do relative clauses come from, how can they be recognised,and what do they depend upon?......Page 351
8.1 Discussion......Page 354
9. Discussion......Page 357
10. Conclusion......Page 359
REFERENCES......Page 361
1. Introduction......Page 366
2. Analysis of nominal anaphors in the test data......Page 367
3. A preference rule for nominal descriptions......Page 368
4. Experimental results......Page 373
5. Implementation......Page 375
REFERENCES......Page 376
1. Introduction......Page 378
2. Conceptual graphs: A brief introduction......Page 379
4. Relevant system components......Page 380
5.1 The main objectives — subject information, coherence and multilinguality......Page 381
5.2 Query mapper — the strategical component......Page 382
5.3.2 Explanation levels......Page 383
5.3.3 Utterance forming......Page 384
5.4 Sample output......Page 385
7. Conclusion......Page 386
REFERENCES......Page 387
V. CORPUS PROCESSING ANDAPPLICATIONS......Page 388
1. Introduction......Page 390
2. Disappointment......Page 391
3. Myth-1: Compositionality of translation......Page 392
4. Myth-2: Possible translation......Page 395
5. Examples: Metonymic nature of language and translation......Page 396
6. Conceptual design of a simple MT system......Page 400
7. Other frameworks and future directions......Page 402
REFERENCES......Page 404
1. Introduction......Page 406
2. F-structure representations......Page 407
3. The mapper......Page 409
4. An example......Page 412
5. Training, testing and performance......Page 414
6. Discussion......Page 416
REFERENCES......Page 417
1. Introduction......Page 418
2.1 Calculation of Word Similarities......Page 419
2.2 Structural matching of parallel sentences......Page 421
3. Experiments of translation rule acquisition......Page 423
3.1 Acquisition of translation rules......Page 424
3.2 The translation rules......Page 425
4. Discussion and Related Works......Page 427
5. Conclusions......Page 428
REFERENCES......Page 429
1. Introduction......Page 430
2. Previous work......Page 431
3. The model......Page 432
4. Results......Page 435
REFERENCES......Page 436
1. Introduction......Page 440
3.1 General estimation of distribution......Page 441
3.2 Variable bag estimation......Page 444
4. Experiments......Page 447
5. Results and observations......Page 449
REFERENCES......Page 450
1. Introduction......Page 452
2.1 Characteristics of Korean......Page 453
2.2 Construction of Hidden Markov Model......Page 454
3.1 Test data......Page 459
3.2 Results......Page 460
REFERENCES......Page 462
1. Introduction......Page 464
2. System overview......Page 465
3. Automata and rules......Page 468
5. Modality synergy......Page 469
6. Architecture of the NL component......Page 471
7. A simple translation example......Page 473
8. Conclusions......Page 474
REFERENCES......Page 475
List and Addresses of Contributors......Page 476
Index of Subjects and Terms......Page 482