Text GenerationCambridge University Press, 1992 M06 26 - 246 páginas This book is concerned with the machine-based generation of natural language text and presents a formal analysis of problems, which in the main have previously only been approached descriptively. In the process of producing discourse, speakers and writers must decide what it is that they want to say and how to present it effectively. Kathleen McKeown's main concern is to identify and formalise principles of discourse so that they can be used in a computational process. The text generation theory she describes has been embodied in a computer program, TEXT, which, given a question, can produce a paragraph length response. An Appendix to the book provides examples of the TEXT system in operation. |
Contenido
Introduction | vii |
12 A processing model | 1 |
13 A sketch of related work | 2 |
14 A text generation theory and method | 4 |
15 System overview | 7 |
16 The database application | 9 |
17 Other issues | 12 |
18 Guide to remaining chapters | 13 |
43 Selection of relevant knowledge | 109 |
432 Comparisons | 110 |
484 Relevancy on the basis of conceptual closeness | 112 |
435 Conclusions | 117 |
44 Schema implementation | 118 |
442 Arc actions | 119 |
444 Graphs used | 120 |
446 The compare and contrast schema | 126 |
Discourse structure | 15 |
21 Rhetorical predicates | 16 |
212 Ordering communicative techniques | 17 |
22 Analysis of texts | 20 |
221 Predicate recursiveness | 26 |
222 Summary of text analysis | 33 |
23 Related research using rhetorical predicates | 34 |
241 Associating technique with purpose | 36 |
25 Selecting a schema | 38 |
26 Filling the schema | 41 |
27 An example | 43 |
28 Future work | 48 |
29 Conclusions | 49 |
Focusing in discourse | 51 |
31 Computational theories and uses of focusing | 52 |
312 Immediate focus | 53 |
321 Global focus and generation | 55 |
322 Immediate focus and generation | 56 |
323 Current focus versus potential focus list | 58 |
324 Current focus versus focus stack | 61 |
325 Other choices | 63 |
326 A focus algorithm for generation | 65 |
328 Overriding the default focus | 66 |
829 The focus algorithm | 67 |
3210 Use of focus sets | 69 |
33 Focus and syntactic structures | 71 |
332 Passing focus information to the tactical component | 73 |
34 Future work | 75 |
35 Conclusions | 76 |
TEXT system implementation | 79 |
41 System components | 80 |
42 Knowledge representation | 83 |
422 Portability | 86 |
423 Summary | 87 |
424 The entityrelationship model | 88 |
425 Use of generalization | 89 |
426 The topic hierarchy | 93 |
427 Relations | 96 |
428 Distinguishing descriptive attributes | 98 |
429 DDAs for database entity generalizations | 99 |
4210 Supporting database attributes | 100 |
4211 Based database attributes | 102 |
4212 DDAs for database entity subsets | 105 |
4213 Constant database attributes | 107 |
45 The tactical component | 129 |
452 The grammar formalism | 130 |
453 A functional grammar | 134 |
454 The unifier | 136 |
455 The TEXT system unifier | 137 |
456 Unifying a sample input with a sample grammar | 140 |
457 Grammar implementation | 143 |
458 Morphology and linearization | 147 |
459 Extensions | 148 |
4511 Advantages | 149 |
46 The dictionary | 151 |
462 Structure of dictionary entries | 152 |
463 General entries | 153 |
464 An example | 156 |
465 Creating the dictionary | 160 |
466 Conclusions | 163 |
47 Practical considerations | 164 |
472 Question coverage | 165 |
Discourse history | 167 |
52 Questions about the difference between entities | 168 |
53 Requests for definitions | 174 |
54 Requests for information | 176 |
55 Summary | 179 |
Related generation research | 181 |
61 Tactical components early systems | 182 |
62 Tactical components later works | 183 |
64 Planning and generation | 185 |
65 Knowledge needed for generation | 186 |
66 Text generation | 187 |
Summary and conclusions | 191 |
73 Discourse coherency | 192 |
75 An evaluation of the generated text | 193 |
76 Limitations of the implemented system | 195 |
77 Future directions | 196 |
772 Relevancy | 199 |
774 User model | 200 |
Sample output of the TEXT system | 201 |
Introduction to Working | 219 |
Resources used | 221 |
Predicate semantics | 223 |
Bibliography | 233 |
Index | 240 |
Términos y frases comunes
aircraft carrier anaphora answer arguments attrs based-db BNKR capabilities are provided catamaran coherency contrast schema cruiser current focus database have REMARKS database systems DB attributes DB-ATTRS default focus definition described DESTROYER determine dictionary discourse goal discourse structure discussion DISPLACEMENT distinguishing descriptive attribute entity-relationship model entry example focus constraints focus stack focusing FUEL CAPACITY FUEL TYPE function global focus grammar guided projectile Hobie Cat identification schema immediate focus implemented input knowledge base knowledge representation lexical linguistic MAXIMUM OPERATING DEPTH MISSILE natural language noun phrase object grammar ocean escort ONR database potential focus list predicate previous discourse proposition selected PROPULSION questions recursive relevant knowledge pool representation schemata semantic sentence grammar shown in Figure SOME-TYPE-OF specified SPEED stack STMTURGRD strategic component sub-type superordinate surface syntactic category tactical component target location TARGET-LOCATION TEXT system topic hierarchy TORPEDOE translation TRAVEL-MODE UNDERWATER verb water-going vehicles WATER-VEHICLE whisky
Referencias a este libro
Handbook of Natural Language Processing Robert Dale,Hermann Moisl,Harold Somers Vista previa limitada - 2000 |
Natural Language Generation: New Results in Artificial Intelligence ... G.A. Kempen Vista de fragmentos - 1987 |