Text Genres and Registers: The Computation of Linguistic Features [recurso electrónico

"This book is a description of some of the most recent advances in text classification as part of a concerted effort to achieve computer understanding of human language. In particular, it addresses state-of-the-art developments in the computation of higher-level linguistic features, ranging from etymology to grammar and syntax for the practical task of text classification according to genres, registers and subject domains. Serving as a bridge between computational methods and sophisticated linguistic analysis, this book will be of particular interest to academics and students of computational linguistics as well as professionals in natural language engineering"--Provided by publisher Preface and Acknowledgements 6 Contents 7 About the Authors 11 Chapter-1 12 Introduction 12 1.1 The Corpus as a Model of Linguistic Use 13 1.2 The Internal and External Dimensions in the Corpus 14 1.3 The Predictive Power of the Corpus 16 1.4 Genres and Registers 17 1.5 Linguistic Variation Across Genres and Registers 20 Chapter-2 21 Language Resources 21 2.1 General Corpora 21 2.1.1 The Brown Corpus and the Brown Family 21 2.1.2 The International Corpus of English (ICE) Family 24 2.1.3 BNC and ANC 27 2.1.3.1 The British National Corpus (BNC) 27 2.1.3.2 The American National Corpus (ANC) 30 2.2 Specialised Collections 32 2.2.1 Wall Street Journal 32 2.2.2 PubMed 33 2.3 Lexical Sources 34 2.3.1 WordNet 34 2.3.2 FrameNet 34 Chapter-3 36 Corpus Annotation and Usable Linguistic Features 36 3.1 Textual Annotation 37 3.2 Grammatical Annotation 38 3.2.1 The LOB Tagset 39 3.2.2 The ICE Tagset 41 3.2.3 A Comparison of LOB and ICE 44 3.3 Syntactic Annotation 48 3.3.1 The Penn Treebank Scheme 49 3.3.2 The ICE Parsing Scheme 51 3.3.3 Summary 53 3.4 Dialogue Act Annotation 53 3.4.1 Notable DA Schemes 56 3.4.2 ISO DA Scheme 58 3.5 Machine Learning and Linguistic Features 60 3.5.1 Machine Learning and Text Classification 60 3.5.2 Weka 62 Chapter-4 64 Etymological Features across Genres and Registers 64 4.1 Research Background 64 4.2 Resources 66 4.2.1 Corpus Resource 66 4.2.2 Lexical Resource 67 4.2.3 Reference Lists 68 4.3 Investigation of Text Categories 69 4.3.1 Descriptive Statistics 69 4.3.2 Borrowed Words and Text Categories 70 4.3.3 Summary 73 4.4 Investigation of Subject Domains 74 4.4.1 Creation of a Sub-corpus 75 4.4.2 Descriptive Statistics 75 4.4.3 Borrowed Words and Domains 76 4.4.4 Summary 78 4.5 Conclusion 79 Chapter-5 80 Part-of-Speech Tags and ICE Text Classification 80 5.1 Research Background 80 5.2 Methodology 81 5.2.1 Experimental Setup 81 5.2.2 Corpus Resources 82 5.2.3 Machine-Learning Tools 83 5.3 Feature Sets 83 5.3.1 Fine-Grained POS Tags (F-POS) 83 5.3.2 BOW 84 5.3.3 Impoverished Tags (I-POS) 84 5.4 Experimental Results 85 5.4.1 Results Obtained From NB Classifier 85 5.4.2 Results Obtained from NB-MN Classifier 86 5.4.3 Discussion 89 5.5 Conclusion 91 Chapter-6 92 Verbs and Text Classification 92 6.1 Transitivity Type and Text Categories 92 6.1.1 The Distribution of Lexical Verbs 93 6.1.1.1 Observations 94 6.1.1.2 Summary 97 6.1.2 The Distribution of Verb Transitivity Types 97 6.1.2.1 ICE-GB Verb Subcategorisations 101 6.1.2.2 Observations 102 6.1.2.3 Summary 104 6.1.3 Conclusion 104 6.2 Infinitive Verbs and Text Categories 106 6.2.1 The Overall Distribution of Infinitives 108 6.2.2 Aux Infinitives 109 6.2.3 Bare Infinitives 112 6.2.4 To-Infinitives 114 6.2.5 For/to-Infinitives 118 6.2.5.1 Object Clauses 120 6.2.5.2 Adverbial Clauses of Result 121 6.2.5.3 Appositional Clauses 121 6.2.5.4 Gerundial Clauses Following NP Introduced by for 123 6.2.6 Summary and Conclusion 124 Chapter-7 125 Adjectives and Text Categories 125 7.1 Adjective and Formality 125 7.1.1 Research Background 125 7.1.2 Methodology 126 7.1.3 Adjective Use across Text Categories 127 7.1.3.1 Adjective Density and Automatic Ranking 127 7.1.3.2 Evaluating Manual and Automatic Rankings 128 7.1.3.3 Linear Regression Analysis 129 7.1.3.4 Adapting to Unseen Data Sets 130 7.1.4 Adjective Density and Automatic Text Classification 132 7.1.5 Conclusion 134 7.2 Adjective Phrase (AJP) and Subject Domains 135 7.2.1 Corpus Resource 135 7.2.1.1 Syntactic Annotation of the ICE-GB 136 7.2.1.2 Term Annotation 136 7.2.2 Investigation of Adjective Use 139 7.2.2.1 The Distribution of AJPs According to Category and Domain 139 7.2.2.2 Use of Adjectives as Linguistic Feature to Classify Texts 140 7.2.3 Conclusion 141 Chapter-8 142 Adverbial Clauses across Text Categories and Registers 142 8.1 Adverbial Clauses across Speech and Writing 143 8.1.1 Adverbial Clauses across Spontaneous and Prepared Speech 144 8.1.2 Adverbial Clauses across Timed and Untimed Essays 145 8.2 Frequency Distribution of Adverbial Subordinators 146 8.3 Discussions and Conclusion 148 Chapter-9 150 Coordination across Modes, Genres and Registers 150 9.1 Methodology and Corpus Data 157 9.2 The Distribution of Coordinators 160 9.3 Syntactic Categories of Coordination Conjoins 163 9.4 Syntactic Functions of Coordination 167 9.5 Conclusion 172 Chapter-10 173 Semantic Features and Authorship Attribution 173 10.1 Corpus Annotated with Ontological Knowledge 177 10.2 Selection and Evaluation of Stylistic Features 181 10.3 Discussions and Conclusion 186 Chapter-11 188 Pragmatics and Dialogue Acts 188 11.1 Corpus Resource 189 11.2 Related Research on the SWBD-DAMSL Scheme 193 11.3 Methodology 195 11.3.1 Machine Learning Techniques 195 11.3.2 Data Preprocessing 195 11.3.3 Research Questions 196 11.4 Classification Results 196 11.5 Qualitative Analysis 199 11.5.1 Hedge 199 11.5.2 Statement-non-Opinion Vs. Statement-Opinion 208 11.5.3 Acknowledge (Backchannel) 211 11.6 Conclusions 219 Chapter-12 221 The Future 221 Appendix A 225 A List of ICE Part-of-Speech Tags 225 Appendix B 232 A List of LOB Part-of-Speech Tags 232 Appendix C 236 A List of Penn Treebank Part-of-Speech Tags 236 Appendix D 238 A List of ICE Parsing Symbols 238 Appendix E 240 A List of Penn Treebank Parsing Symbols 240 Appendix F 241 A List of Adverbial Subordinators in Speech 241 Appendix G 244 A List of Adverbial Subordinators in Writing 244 Bibliography 246 Index 262 "This book is a description of some of the most recent advances in text classification as part of a concerted effort to achieve computer understanding of human language. In particular, it addresses state-of-the-art developments in the computation of higher-level linguistic features, ranging from etymology to grammar and syntax for the practical task of text classification according to genres, registers and subject domains. Serving as a bridge between computational methods and sophisticated linguistic analysis, this book will be of particular interest to academics and students of computational linguistics as well as professionals in natural language engineering"--Jacket Front Matter....Pages i-xiii Introduction....Pages 1-9 Language Resources....Pages 11-25 Corpus Annotation and Usable Linguistic Features....Pages 27-54 Etymological Features Across Genres and Registers....Pages 55-70 Part-of-Speech Tags and ICE Text Classification....Pages 71-82 Verbs and Text Classification....Pages 83-115 Adjectives and Text Categories....Pages 117-133 Adverbial Clauses Across Text Categories and Registers....Pages 135-142 Coordination Across Modes, Genres and Registers....Pages 143-165 Semantic Features and Authorship Attribution....Pages 167-181 Pragmatics and Dialogue Acts....Pages 183-215 The Future....Pages 217-220 Back Matter....Pages 221-267