Computing Attitude and Affect in Text: Theory and Applications (eBook)

James G. Shanahan, Yan Qu, Janyce Wiebe (Herausgeber)

eBook Download: PDF

2006 | 2006
XVI, 341 Seiten
Springer Netherland (Verlag)
978-1-4020-4102-0 (ISBN)

Infos

Human Language Technology (HLT) and Natural Language Processing (NLP) systems have typically focused on the 'factual' aspect of content analysis. Other aspects, including pragmatics, opinion, and style, have received much less attention. However, to achieve an adequate understanding of a text, these aspects cannot be ignored. The chapters in this book address the aspect of subjective opinion, which includes identifying different points of view, identifying different emotive dimensions, and classifying text by opinion. Various conceptual models and computational methods are presented. The models explored in this book include the following: distinguishing attitudes from simple factual assertions; distinguishing between the author's reports from reports of other people's opinions; and distinguishing between explicitly and implicitly stated attitudes. In addition, many applications are described that promise to benefit from the ability to understand attitudes and affect, including indexing and retrieval of documents by opinion; automatic question answering about opinions; analysis of sentiment in the media and in discussion groups about consumer products, political issues, etc. ; brand and reputation management; discovering and predicting consumer and voting trends; analyzing client discourse in therapy and counseling; determining relations between scientific texts by finding reasons for citations; generating more appropriate texts and making agents more believable; and creating writers' aids. The studies reported here are carried out on different languages such as English, French, Japanese, and Portuguese. Difficult challenges remain, however. It can be argued that analyzing attitude and affect in text is an 'NLP'-complete problem.

---

Human Language Technology (HLT) and Natural Language Processing (NLP) systems have typically focused on the "e;factual"e; aspect of content analysis. Other aspects, including pragmatics, opinion, and style, have received much less attention. However, to achieve an adequate understanding of a text, these aspects cannot be ignored. The chapters in this book address the aspect of subjective opinion, which includes identifying different points of view, identifying different emotive dimensions, and classifying text by opinion. Various conceptual models and computational methods are presented. The models explored in this book include the following: distinguishing attitudes from simple factual assertions; distinguishing between the author's reports from reports of other people's opinions; and distinguishing between explicitly and implicitly stated attitudes. In addition, many applications are described that promise to benefit from the ability to understand attitudes and affect, including indexing and retrieval of documents by opinion; automatic question answering about opinions; analysis of sentiment in the media and in discussion groups about consumer products, political issues, etc. ; brand and reputation management; discovering and predicting consumer and voting trends; analyzing client discourse in therapy and counseling; determining relations between scientific texts by finding reasons for citations; generating more appropriate texts and making agents more believable; and creating writers' aids. The studies reported here are carried out on different languages such as English, French, Japanese, and Portuguese. Difficult challenges remain, however. It can be argued that analyzing attitude and affect in text is an "e;NLP"e;-complete problem.

Inhalt

TABLE OF CONTENTS 6
Preface 12
Contextual Valence Shifters 18
1. Introduction 18
2. From Simple Valence to Contextually Determined Valence 19
3. Contextual Valence Shifters 20
4. Conclusion 26
5. Bibliography 27
Conveying Attitude with Reported Speech 28
1. Introduction 28
2. Evidential Analysis of Reported Speech 29
3. Profile Structure 31
4. Extended Example 33
5. Source List Annotation 34
6. Extension to Other Attribution 37
7. Conclusion 37
8. Acknowledgements 38
9. Bibliography 38
Where Attitudinal Expressions Get their Attitude 40
1. Research Questions to Motivate the Study of Attitudinal Expressions 40
2. Starting Points – Prototypical Attitudinal Expressions 41
3. Text Topicality: Players 41
4. Text Topicality: Moves 42
5. Identifying Players 42
6. The Case for Animacy: Adjectival Attributes and Genitive Attributes 42
7. The Case for Syntactic Structure: Situational Reference 43
8. Using Syntactic Patterns more Systematically 45
9. Generalizing from Syntactic Patterns to the Lexicon 46
10. Conclusions 46
11. Bibliography 47
The Subjectivity of Lexical Cohesion in Text 56
1. Introduction 56
2. Theoretical Background 57
3. Experimental Study 58
4. Discussion 60
5. Acknowledgements 61
6. Bibliography 61
A Weighted Referential Activity Dictionary 63
1. Introduction 64
2. Methods 66
3. Results 72
4. Bibliography 74
Certainty Identification in Texts: Categorization Model and Manual Tagging Results 75
1. Analytical Framework 76
2. Proposed Certainty Categorization Model 79
3. Empirical Study 82
4. Applications 88
5. Conclusions and Future Work 88
6. Acknowledgements 89
7. Bibliography 89
Evaluating an Opinion Annotation Scheme Using a New Multi-Perspective Question and Answer Corpus 91
1. Introduction 91
2. Low-Level Perspective Information 92
3. The MPQA NRRC Corpus 94
4. Multi-Perspective Question and Answer Corpus Creation 94
5. Evaluation of Perspective Annotations for MPQA 97
6. Conclusions and Future Work 103
7. Acknowledgements 103
8. Bibliography 104
A Computational Semantic Lexicon of French Verbs of Emotion 121
1. Introduction 121
2. Semantic Lexicon Description 121
3. FEELING System 128
4. Evaluation 134
5. Related Work 135
6. Conclusion 135
7. Bibliography 135
Extracting Opinion Propositions and Opinion Holders using Syntactic and Lexical Cues 137
1. Introduction 137
2. Data 139
3. Opinion-Oriented Words 142
4. Identifying Opinion Propositions 144
5. Results 148
6. Error Analysis 150
7. Discussion 151
8. Acknowledgments 152
9. Bibliography 152
Approaches for Automatically Tagging Affect: Steps Toward an Effective and Efficient Tool 154
1. Introduction 154
2. Background 155
3. Rochester Marriage-Counseling Corpus 156
4. Approaches to Tagging 157
5. Evaluations 164
6. Discussion 165
7. CATS Tool 167
8. Related Work 168
9. Conclusion 168
10. Acknowledgments 169
11. Bibliography 169
Argumentative Zoning for Improved Citation Indexing 170
1. Citation Indexing and Citation Maps 170
2. Argumentative Zoning and Author Affect 172
3. Meta-discourse 174
4. Human Annotation of Author Affect 176
5. Features for Author Affect 178
6. Evaluation 178
7. Conclusion 179
8. Bibliography 179
Two Exploratory Studies Politeness and Bias in Dialogue Summarization 181
1. Introduction 182
2. First Study: Politeness and Bias in Unconstrained Dialogue Summarization 184
3. Second Study: Politeness and Bias in Constrained Dialogue Summarization 188
4. Comparison 190
5. Conclusion and Outlook 191
6. Acknowledgements 192
7. Bibliography 192
8. Appendix I 195
Generating More-Positive and More-Negative Text 196
1. Near-Synonyms and Attitudinal Nuances 196
2. Related Work 198
3. Estimating the Relative Semantic Orientation of Text 198
4. Word Sense Disambiguation 199
5. Analysis 199
6. Generation 200
7. Experiments 201
8. Evaluation 204
9. Conclusion 205
10. Acknowledgements 205
11. Bibliography 205
Identifying Interpersonal Distance using Systemic Features 208
1. Introduction 209
2. Systemic Functional Linguistics 209
3. Representing System Networks 213
4. Identifying Registers 218
5. Conclusion 221
6. Bibliography 221
Corpus-Based Study of Scientific Methodology: Comparing the Historical and Experimental Sciences 224
1. Introduction 225
2. Background 225
3. Systemic Indicators as Textual Features 228
4. Experimental Study 231
5. Example Texts 236
6. Conclusions 237
7. Acknowledgements 237
8. References 237
Argumentative Zoning Applied to Critiquing Novices™ Scientific Abstracts 241
1. Introduction 242
2. The SciPo System 242
3. Argumentative Zoning for Portuguese Texts 245
4. Evaluation of SciPo™s Critiquing Tool 250
5. Conclusions 252
6. Acknowledgements 252
7. Bibliography 252
Using Hedges to Classify Citations in Scientific Articles 255
1. Scientific Writing, the Need for Affect, and Its Role in Citation Analysis 255
2. Hedging in Scientific Writing 256
3. Classifying Citations in Scientific Writing 258
4. Determining the Importance of Hedges in Citation Contexts 260
5. A Citation Indexing Tool for Biomedical Literature Analysis 264
6. Conclusions and Future Work 269
7. Acknowledgements 269
8. Bibliography 269
Towards a Robust Metric of Polarity 272
1. Introduction 272
2. Related Work 273
3. Classes of Polar Expression 275
4. Recognizing Polar Language 276
5. Topic Detection in Online Messages 277
6. The Intersection of Topic and Polarity 279
7. Empirical Analysis 280
8. Metrics for Topic and Polarity 282
9. Conclusions and Future Work 284
10. References 285
Characterizing Buzz and Sentiment in Internet Sources: Linguistic Summaries and Predictive Behaviors 287
1. Introduction and Motivation 287
2. Linguistic Summaries 288
3. Example Applications 295
4. TRENDS-2™ Infrastructure 298
5. Previous and Related Work 299
6. Open R& D and Application Issues
7. Bibliography 300
Good News or Bad News? Let the Market Decide 303
1. Introduction 303
2. Experiments 304
3. Results 305
4. Conclusions 306
Opinion Polarity Identification of Movie Reviews 308
1. Introduction 308
2. Related Research 309
3. Probabilistic Approaches to Polarity Identification 310
4. Features for Analysis 311
5. Part of Speech Feature Selection 312
6. Experiments 313
7. Synonymy and Hypernymy Feature Generalization 317
8. Selection by Ranking 319
9. Discussion 319
10. Conclusion 320
11. Acknowledgements 320
12. Bibliography 320
Multi-Document Viewpoint Summarization Focused on Facts, Opinion and Knowledge 322
1. Introduction 323
2. Experiment Overview: Multi-Document Viewpoint Summarization with Summary Types 324
3. Sentence-type Annotation 328
4. Genre Classification 330
5. Experiment Results 333
6. Conclusion 338
7. Acknowledgement 339
8. Bibliography 339
INDEX 342

Produkt-Details

• Erscheint lt. Verlag: 17.1.2006
• Reihe/Serie: The Information Retrieval Series
• Reihe/Serie: The Information Retrieval Series
• Zusatzinfo: XVI, 341 p.
• Verlagsort: Dordrecht
• Sprache: englisch
• Themenwelt: Geisteswissenschaften ► Sprach- / Literaturwissenschaft
• Themenwelt: Mathematik / Informatik ► Informatik ► Datenbanken
• Themenwelt: Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik
• Themenwelt: Sozialwissenschaften ► Kommunikation / Medien ► Buchhandel / Bibliothekswesen
• Schlagworte: Artificial Intelligence • berck • Corpus • Intelligence • Natural Language Processing • Text • therapy
• ISBN-10: 1-4020-4102-0 / 1402041020
• ISBN-13: 978-1-4020-4102-0 / 9781402041020