+353-1-416-8900REST OF WORLD
+44-20-3973-8888REST OF WORLD
1-917-300-0470EAST COAST U.S
1-800-526-8630U.S. (TOLL FREE)

Automatic Detection of Irony. Opinion Mining in Microblogs and Social Media. Edition No. 1

  • Book

  • 210 Pages
  • October 2019
  • John Wiley and Sons Ltd
  • ID: 5838339
In recent years, there has been a proliferation of opinion-heavy texts on the Web: opinions of Internet users, comments on social networks, etc. Automating the synthesis of opinions has become crucial to gaining an overview on a given topic. Current automatic systems perform well on classifying the subjective or objective character of a document. However, classifications obtained from polarity analysis remain inconclusive, due to the algorithms' inability to understand the subtleties of human language. Automatic Detection of Irony presents, in three stages, a supervised learning approach to predicting whether a tweet is ironic or not. The book begins by analyzing some everyday examples of irony and presenting a reference corpus. It then develops an automatic irony detection model for French tweets that exploits semantic traits and extralinguistic context. Finally, it presents a study of portability in a multilingual framework (Italian, English, Arabic).

Table of Contents

Preface ix

Introduction xi

Chapter 1. From Opinion Analysis to Figurative Language Treatment 1

1.1. Introduction 1

1.2. Defining the notion of opinion 3

1.2.1. The many faces of opinion 3

1.2.2. Opinion as a structured model 4

1.2.3. Opinion extraction: principal approaches 5

1.3. Limitations of opinion analysis systems 7

1.3.1. Opinion operators 8

1.3.2. Domain dependency 9

1.3.3. Implicit opinions 10

1.3.4. Opinions and discursive context above phrase level 11

1.3.5. Presence of figurative expressions 12

1.4. Definition of figurative language 13

1.4.1. Irony 13

1.4.2. Sarcasm 18

1.4.3. Satire 20

1.4.4. Metaphor 21

1.4.5. Humor 22

1.5. Figurative language: a challenge for NLP 23

1.6. Conclusion 23

Chapter 2. Toward Automatic Detection of Figurative Language 25

2.1. Introduction 25

2.2. The main corpora used for figurative language 27

2.2.1. Corpora annotated for irony/sarcasm 28

2.2.2. Corpus annotated for metaphors 33

2.3. Automatic detection of irony, sarcasm and satire 36

2.3.1. Surface and semantic approaches 36

2.3.2. Pragmatic approaches 39

2.4. Automatic detection of metaphor 51

2.4.1. Surface and semantic approaches 52

2.4.2. Pragmatic approaches 53

2.5. Automatic detection of comparison 58

2.6. Automatic detection of humor 58

2.7. Conclusion 61

Chapter 3. A Multilevel Scheme for Irony Annotation in Social Network Content 63

3.1. Introduction 63

3.2. The FrIC 65

3.3. Multilevel annotation scheme 66

3.3.1. Methodology 66

3.3.2. Annotation scheme 69

3.4. The annotation campaign 79

3.4.1. Glozz 79

3.4.2. Data preparation 80

3.4.3. Annotation procedure 81

3.5. Results of the annotation campaign 83

3.5.1. Qualitative results 83

3.5.2. Quantitative results 84

3.5.3. Correlation between different levels of the annotation scheme 89

3.6. Conclusion 93

Chapter 4. Three Models for Automatic Irony Detection 95

4.1. Introduction 95

4.2. The FrICAuto corpus 97

4.3. The SurfSystem model: irony detection based on surface features 99

4.3.1. Selected features 99

4.3.2. Experiments and results 101

4.4. The PragSystem model: irony detection based on internal contextual features 104

4.4.1. Selected features 104

4.4.2. Experiments and results 109

4.4.3. Discussion 116

4.5. The QuerySystem model: developing a pragmatic contextual approach for automatic irony detection 118

4.5.1. Proposed approach 118

4.5.2. Experiments and results 122

4.5.3. Evaluation of the query-based method 123

4.6. Conclusion 124

Chapter 5. Towards a Multilingual System for Automatic Irony Detection 127

5.1. Introduction 127

5.2. Irony in Indo-European languages 128

5.2.1. Corpora 128

5.2.2. Results of the annotation process 130

5.2.3. Summary 139

5.3. Irony in Semitic languages 140

5.3.1. Specificities of Arabic 142

5.3.2. Corpus and resources 143

5.3.3. Automatic detection of irony in Arabic tweets 146

5.4. Conclusion 149

Conclusion 151

Appendix 155

References 169

Index 189

Authors

Jihen Karoui Farah Benamara Veronique Moriceau