Computing Attitude and Affect in Text: Theory and ApplicationsJames G. Shanahan, Yan Qu, Janyce Wiebe Springer Science & Business Media, 22 nov. 2005 - 341 pages 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. |
Table des matières
Contextual Valence Shifters | 1 |
2 From Simple Valence to Contextually Determined Valence | 2 |
3 Contextual Valence Shifters | 3 |
4 Conclusion | 9 |
Conveying Attitude with Reported Speech | 11 |
2 Evidential Analysis of Reported Speech | 12 |
3 Profile Structure | 14 |
4 Extended Example | 16 |
1 Introduction | 172 |
Politeness and Bias in Unconstrained Dialogue Summarization | 174 |
Politeness and Bias in Constrained Dialogue Summarization | 178 |
4 Comparison | 180 |
5 Conclusion and Outlook | 181 |
Generating MorePositive and MoreNegative Text | 187 |
2 Related Work | 189 |
4 Word Sense Disambiguation | 190 |
5 Source List Annotation | 17 |
6 Extension to Other Attribution | 20 |
Where Attitudinal Expressions Get their Attitude | 23 |
2 Starting Points Prototypical Attitudinal Expressions | 24 |
Moves | 25 |
Situational Reference | 26 |
8 Using Syntactic Patterns more Systematically | 28 |
9 Generalizing from Syntactic Patterns to the Lexicon | 29 |
Analysis of Linguistic Features Associated with Point of View for Generating Stylistically Appropriate Text | 33 |
2 Perspectives in Corpus | 34 |
3 Associated Features | 36 |
4 Implications for Natural Language Generation and Automatic Recognition of Point of View | 38 |
The Subjectivity of Lexical Cohesion in Text | 41 |
2 Theoretical Background | 42 |
3 Experimental Study | 43 |
4 Discussion | 45 |
A Weighted Referential Activity Dictionary | 49 |
1 Introduction | 50 |
2 Methods | 52 |
3 Results | 58 |
Certainty Identification in Texts Categorization Model and Manual Tagging Results | 61 |
1 Analytical Framework | 62 |
2 Proposed Certainty Categorization Model | 65 |
3 Empirical Study | 68 |
4 Applications | 74 |
Evaluating an Opinion Annotation Scheme Using a New MultiPerspective Question and Answer Corpus | 77 |
2 LowLevel Perspective Information | 78 |
3 The MPQA NRRC Corpus | 80 |
5 Evaluation of Perspective Annotations for MPQA | 83 |
6 Conclusions and Future Work | 89 |
Validating the Coverage of Lexical Resources for Affect Analysis and Automatically Classifying New Words along Semantic Axes | 93 |
1 Introduction | 94 |
2 The Current Clairvoyance Affect Lexicon | 95 |
3 Emotive Patterns | 97 |
4 Scoring the Intensity of Candidate Affect Words | 101 |
5 Future Work | 105 |
6 Conclusions | 106 |
A Computational Semantic Lexicon of French Verbs of Emotion | 109 |
3 FEELING System | 116 |
4 Evaluation | 122 |
5 Related Work | 123 |
Extracting Opinion Propositions and Opinion Holders using Syntactic and Lexical Cues | 125 |
2 Data | 127 |
3 OpinionOriented Words | 130 |
4 Identifying Opinion Propositions | 132 |
5 Results | 136 |
6 Error Analysis | 138 |
7 Discussion | 139 |
Approaches for Automatically Tagging Affect | 143 |
2 Background | 144 |
3 Rochester MarriageCounseling Corpus | 145 |
4 Approaches to Tagging | 146 |
5 Evaluations | 153 |
6 Discussion | 154 |
7 CATS Tool | 156 |
8 Related Work | 157 |
Argumentative Zoning for Improved Citation Indexing | 159 |
2 Argumentative Zoning and Author Affect | 161 |
3 Metadiscourse | 163 |
4 Human Annotation of Author Affect | 165 |
5 Features for Author Affect | 167 |
7 Conclusion | 168 |
Politeness and Bias in Dialogue Summarization Two Exploratory Studies | 171 |
6 Generation | 191 |
7 Experiments | 192 |
8 Evaluation | 195 |
9 Conclusion | 196 |
Identifying Interpersonal Distance using Systemic Features | 199 |
1 Introduction | 200 |
3 Representing System Networks | 204 |
4 Identifying Registers | 209 |
5 Conclusion | 212 |
CorpusBased Study of Scientific Methodology Comparing the Historical and Experimental Sciences | 215 |
1 Introduction | 216 |
3 Systemic Indicators as Textual Features | 219 |
4 Experimental Study | 222 |
5 Example Texts | 227 |
6 Conclusions | 228 |
Argumentative Zoning Applied to Critiquing Novices Scientific Abstracts | 233 |
1 Introduction | 234 |
3 Argumentative Zoning for Portuguese Texts | 237 |
4 Evaluation of SciPos Critiquing Tool | 242 |
5 Conclusions | 244 |
Using Hedges to Classify Citations in Scientific Articles | 247 |
2 Hedging in Scientific Writing | 248 |
3 Classifying Citations in Scientific Writing | 250 |
4 Determining the Importance of Hedges in Citation Contexts | 252 |
5 A Citation Indexing Tool for Biomedical Literature Analysis | 256 |
6 Conclusions and Future Work | 261 |
Towards a Robust Metric of Polarity | 265 |
2 Related Work | 266 |
3 Classes of Polar Expression | 268 |
4 Recognizing Polar Language | 269 |
5 Topic Detection in Online Messages | 270 |
6 The Intersection of Topic and Polarity | 272 |
7 Empirical Analysis | 273 |
8 Metrics for Topic and Polarity | 275 |
9 Conclusions and Future Work | 277 |
Characterizing Buzz and Sentiment in Internet Sources Linguistic Summaries and Predictive Behaviors | 281 |
2 Linguistic Summaries | 282 |
3 Example Applications | 289 |
4 TRENDS2 Infrastructure | 292 |
5 Previous and Related Work | 293 |
Good News or Bad News? Let the Market Decide | 297 |
2 Experiments | 298 |
3 Results | 299 |
4 Conclusions | 300 |
Opinion Polarity Identification of Movie Reviews | 303 |
2 Related Research | 304 |
3 Probabilistic Approaches to Polarity Identification | 305 |
4 Features for Analysis | 306 |
5 Part of Speech Feature Selection | 307 |
6 Experiments | 308 |
7 Synonymy and Hypernymy Feature Generalization | 312 |
8 Selection by Ranking | 314 |
10 Conclusion | 315 |
MultiDocument Viewpoint Summarization Focused on Facts Opinion and Knowledge | 317 |
1 Introduction | 318 |
MultiDocument Viewpoint Summarization with Summary Types | 319 |
3 Sentencetype Annotation | 323 |
4 Genre Classification | 325 |
5 Experiment Results | 328 |
6 Conclusion | 333 |
337 | |
Autres éditions - Tout afficher
Computing Attitude and Affect in Text: Theory and Applications James G. Shanahan,Yan Qu,Janyce Wiebe Aucun aperçu disponible - 2014 |
Computing Attitude and Affect in Text: Theory and Applications James G. Shanahan,Yan Qu,Janyce Wiebe Aucun aperçu disponible - 2009 |
Expressions et termes fréquents
abstract adjectives affect words algorithm analysis answer approach Argumentative Zoning attitude attitudinal attribute automatic behaviour bigrams certainty markers chapter CiteSeer classifier Computational Linguistics context corpus cross-validation data set database described dialogue dictionary discourse document Email emotion evaluation example expressed extraction feature selection Figure FrameNet frequency genre groups Hatzivassiloglou hedging cues hypernym identified interaction interpersonal distance labeled lexical lexical semantic lexicon linguistic summary machine learning meaning metafunction methods modal multi-document summarization n-gram Natural Language Processing near-synonyms negative nouns opinion questions opinion words OPINION-HOLDER opinion-oriented OPINION-PROPOSITION paper patterns polarity positive precision Precision Recall PropBank propositional opinions reported speech representation retrieval rhetorical score segments semantic orientation sentence types structure subjects Support Vector Machine synsets syntactic Table tagging techniques Teufel text segment topic unigrams valence vector verbs weight Wiebe WordNet WRAD
Fréquemment cités
Page 316 - Proceedings of the Thirty-Fifth Annual Meeting of the Association for Computational Linguistics and Eighth Conference of the European Chapter of the Association for Computational Linguistics, 1 997. [52] Krebs B., "E-mail Scam Sought to defraud PayPal customers", Newshytes (19 December, 2001), http://www.newsbytes.com/news/01/l73120.html.