词义消歧：算法与应用=WORD SENSE DISAMBIGUATION:Algorithms and ApplicationsPDF|Epub|txt|kindle电子书版本网盘下载

词义消歧：算法与应用=WORD SENSE DISAMBIGUATION:Algorithms and ApplicationsPDF格式电子书版下载

注意：本站所有压缩包均有解压码： 点击下载压缩包解压工具

1 Introduction&Eneko Agirre and Philip Edmonds1

1.1 Word Sense Disambiguation1

1.2 ABrief History of WSD Research4

1.3 What is a Word Sense?8

1.4 Applications of WSD10

1.5 Basic Approaches to WSD12

1.6 State-of-the-Art Performance14

1.7 Promising Directions15

1.8 Overview of This Book19

1.9 Further Reading21

References22

2 Word Senses&Adam Kilgarriff29

2.1 Introduction29

2.2 Lexicographers30

2.3 Philosophy32

2.3.1 Meaning is Something You Do32

2.3.2 The Fregean Tradition and Reification33

2.3.3 Two Incompatible Semantics?33

2.3.4 Implications for Word Senses34

2.4 Lexicalization35

2.5 Corpus Evidence39

2.5.1 Lexicon Size41

2.5.2 Quotations42

2.6 Conclusion43

2.7 Further Reading44

Acknowledgments45

References45

3 Making Sense About Sense&Nancy Ide and Yorick Wilks47

3.1 Introduction47

3.2 WSD and the Lexicographers49

3.3 WSD and Sense Inventories51

3.4 NLP Applications and WSD55

3.5 What Level of Sense Distinctions Do We Need for NLP,If Any?58

3.6 What Now for WSD?64

3.7 Conclusion68

References68

4 Evaluation of WSD Systems&Martha Palmer,Hwee Tou Ng and Hoa Trang Dang75

4.1 Introduction75

4.1.1 Terminology76

4.1.2 Overview80

4.2 Background81

4.2.1 Word Net and Semcor81

4.2.2 The Line and Interest Corpora83

4.2.3 The DSO Corpus84

4.2.4 Open Mind Word Expert85

4.3 Evaluation Using Pseudo-Words86

4.4 Senseval Evaluation Exercises86

4.4.1 Senseval-187

Evaluation and Scoring88

4.4.2 Senseval-288

English All-Words Task89

English Lexical Sample Task89

4.4.3 Comparison of Tagging Exercises91

4.5 Sources of Inter-Annotator Disagreement92

4.6 Granularity of Sense:Groupings for WordNet95

4.6.1 Criteria for WordNet Sense Grouping96

4.6.2Analysis of Sense Grouping97

4.7 Senseval-398

4.8 Discussion99

References102

5 Knowledge-Based Methods for WSD&Rada Mihalcea107

5.1 Introduction107

5.2 Lesk Algorithm108

5.2.1 Variations of the Lesk Algorithm110

Simulated Annealing110

Simplified Lesk Algorithm111

Augmented Semantic Spaces113

Summary113

5.3 Semantic Similarity114

5.3.1 Measures of Semantic Similarity114

5.3.2 Using Semantic Similarity Within a Local Context117

5.3.3 Using Semantic Similarity Within a Global Context118

5.4 Selectional Preferences119

5.4.1 Preliminaries:Learning Word-to-Word Relations120

5.4.2 Learning Selectional Preferences120

5.4.3 Using Selectional Preferences122

5.5 Heuristics for Word Sense Disambiguation123

5.5.1 Most Frequent Sense123

5.5.2 One Sense Per Discourse124

5.5.3 One Sense Per Collocation124

5.6 Knowledge-Based Methods at Senseval-2125

5.7 Conclusions126

References127

6 Unsupervised Corpus-Based Methods for WSD&Ted Pedersen133

6.1 Introduction133

6.1.1 Scope134

6.1.2 Motivation136

Distributional Methods137

Translational Equivalence139

6.1.3 Approaches140

6.2 Type-Based Discrimination141

6.2.1 Representation of Context142

6.2.2 Algorithms145

Latent Semantic Analysis（LSA）146

Hyperspace Analogue to Language（HAL）147

Clustering By Committee（CBC）148

6.2.3 Discussion150

6.3 Token-Based Discrimination150

6.3.1 Representation of Context151

6.3.2 Algorithms151

Context Group Discrimination152

McQuitty's Similarity Analysis154

6.3.3 Discussion157

6.4 Translational Equivalence158

6.4.1 Representation of Context159

6.4.2 Algorithms159

6.4.3 Discussion160

6.5 Conclusions and the Way Forward161

Acknowledgments162

References162

7 Supervised Corpus-Based Methods for WSD&Llu？s Màrquez,Gerard Escudero,David Martínez and German Rigau167

7.1 Introduction to Supervised WSD167

7.1.1 Machine Learning for Classification168

An Example on WSD170

7.2 A Survey of Supervised WSD171

7.2.1 Main Corpora Used172

7.2.2 Main Sense Repositories173

7.2.3 Representation of Examples by Means of Features174

7.2.4 Main Approaches to Supervised WSD175

Probabilistic Methods175

Methods Based on the Similarity of the Examples176

Methods Based on Discriminating Rules177

Methods Based on Rule Combination179

Linear Classifiers and Kernel-Based Approaches179

Discourse Properties:The Yarowsky Bootstrapping Algorithm181

7.2.5 Supervised Systems in the Senseval Evaluations183

7.3 An Empirical Study of Supervised Algorithms for WSD184

7.3.1 Five Learning Algorithms Under Study185

Naive Bayes （NB）185

Exemplar-Based Learning（kNN）186

Decision Lists （DL）187

AdaBoost（AB）187

Support Vector Machines（SVM）189

7.3.2 Empirical Evaluation on the DSO Corpus190

Experiments191

7.4 Current Challenges of the Supervised Approach195

7.4.1 Right-Sized Training Sets195

7.4.2 Porting Across Corpora196

7.4.3 The Knowledge Acquisition Bottleneck197

Automatic Acquisition of Training Examples198

Active Learning199

Combining Training Examples from Different Words199

Parallel Corpora200

7.4.4 Bootstrapping201

7.4.5 Feature Selection and Parameter Optimization202

7.4.6 Combination of Algorithms and Knowledge Sources203

7.5 Conclusions and Future Trends205

Acknowledgments206

References207

8 Knowledge Sources for WSD&Eneko Agirre and Mark Stevenson217

8.1 Introduction217

8.2 Knowledge Sources Relevant to WSD218

8.2.1 Syntactic219

Part of Speech （KS 1）219

Morphology（KS 2）219

Collocations（KS 3）220

Subcategorization（KS 4）220

8.2.2 Semantic220

Frequency of Senses（KS 5）220

Semantic Word Associations（KS 6）221

Selectional Preferences（KS 7）221

Semantic Roles（KS 8）222

8.2.3 Pragmatic/Topical222

Domain（KS 9）222

Topical Word Association（KS 10）222

Pragmatics（KS 11）223

8.3 Features and Lexical Resources223

8.3.1 Target-Word Specific Features224

8.3.2 Local Features225

8.3.3 Global Features227

8.4 Identifying Knowledge Sources inActual Systems228

8.4.1 Senseval-2 Systems229

8.4.2 Senseval-3 Systems231

8.5 Comparison of Experimental Results231

8.5.1 Senseval Results232

8.5.2 Yarowsky and Florian（2002）233

8.5.3 Lee andNg （2002）234

8.5.4 Martínez et al.（2002）237

8.5.5 Agirre and Martínez（2001 a）238

8.5.6 Stevenson and Wilks（2001）240

8.6 Discussion242

8.7 Conclusions245

Acknowledgments246

References247

9 Automatic Acquisition of Lexical Information and Examples&Julio Gonzalo and Felisa Verdejo253

9.1 Introduction253

9.2 Mining Topical Knowledge About Word Senses254

9.2.1 Topic Signatures255

9.2.2 Association of Web Directories to Word Senses257

9.3 Automatic Acquisition of Sense-Tagged Corpora258

9.3.1 Acquisition by Direct Web Searching258

9.3.2 Bootstrapping from Seed Examples261

9.3.3 Acquisition via Web Directories263

9.3.4 Acquisition via Cross-Language Evidence264

9.3.5 Web-Based Cooperative Annotation268

9.4 Discussion269

Acknowledgments271

References272

10 Domain-Specific WSD&Paul Buitelaar,Bernardo Magnini,Carlo Strapparava and Piek Vossen275

10.1 Introduction275

10.2 Approaches to Domain-Specific WSD277

10.2.1 Subject Codes277

10.2.2 Topic Signatures and Topic Variation282

Topic Signatures282

Topic Variation283

10.2.3 Domain Tuning284

Top-down Domain Tuning285

Bottom-up Domain Tuning285

10.3 Domain-Specific Disambiguation in Applications288

10.3.1 User-Modeling forRecommender Systems288

10.3.2 Cross-Lingual Information Retrieval289

10.3.3 The MEANING Project292

10.4 Conclusions295

References296

11 WSD in NLP Applications&Philip Resnik299

11.1 Introduction299

11.2 Why WSD?300

Argument from Faith300

Argument by Analogy301

Argument from Specific Applications302

11.3 Traditional WSD in Applications303

11.3.1 WSD in Traditional Information Retrieval304

11.3.2 WSD in Applications Related to Information Retrieval307

Cross-Language IR308

Question Answering309

Document Classification312

11.3.3 WSD in Traditional Machine Translation313

11.3.4 Sense Ambiguity in Statistical Machine Translation315

11.3.5 Other Emerging Applications317

11.4 Alternative Conceptions of Word Sense320

11.4.1 Richer Linguistic Representations320

11.4.2 Patterns of Usage321

11.4.3 Cross-Language Relationships323

11.5 Conclusions325

Acknowledgments325

References326

A Resources for WSD339

A.1 Sense Inventories339

A.1.1 Dictionaries339

A.1.2 Thesauri341

A.1.3 Lexical Knowledge Bases341

A.2 Corpora343

A.2.1 Raw Corpora343

A.2.2 Sense-Tagged Corpora345

A.2.3 Automatically Tagged Corpora347

A.3 Other Resources348

A.3.1 Software348

A.3.2 Utilities,Demos,and Data349

A.3.3 Language Data Providers350

A.3.4 Organizations and Mailing Lists350

Index of Terms353

Index of Authors and Algorithms361