Word Sense Disambiguation: Algorithms and Applications

Front Cover
Eneko Agirre, Philip Edmonds
Springer Science & Business Media, Nov 16, 2007 - Language Arts & Disciplines - 366 pages
0 Reviews
Graeme Hirst University of Toronto Of the many kinds of ambiguity in language, the two that have received the most attention in computational linguistics are those of word senses and those of syntactic structure, and the reasons for this are clear: these ambiguities are overt, their resolution is seemingly essential for any prac- cal application, and they seem to require a wide variety of methods and knowledge-sources with no pattern apparent in what any particular - stance requires. Right at the birth of artificial intelligence, in his 1950 paper “Computing machinery and intelligence”, Alan Turing saw the ability to understand language as an essential test of intelligence, and an essential test of l- guage understanding was an ability to disambiguate; his example involved deciding between the generic and specific readings of the phrase a winter’s day. The first generations of AI researchers found it easy to construct - amples of ambiguities whose resolution seemed to require vast knowledge and deep understanding of the world and complex inference on this kno- edge; for example, Pharmacists dispense with accuracy. The disambig- tion problem was, in a way, nothing less than the artificial intelligence problem itself. No use was seen for a disambiguation method that was less than 100% perfect; either it worked or it didn’t. Lexical resources, such as they were, were considered secondary to non-linguistic common-sense knowledge of the world.
 

What people are saying - Write a review

We haven't found any reviews in the usual places.

Contents

1 Introduction
1
12 A Brief History of WSD Research
4
13 What is a Word Sense?
8
14 Applications of WSD
10
15 Basic Approaches to WSD
12
16 StateoftheArt Performance
14
17 Promising Directions
15
18 Overview of This Book
19
The Yarowsky Bootstrapping Algorithm
181
725 Supervised Systems in the Senseval Evaluations
183
73 An Empirical Study of Supervised Algorithms for WSD
184
731 Five Learning Algorithms Under Study
185
ExemplarBased Learning kNN
186
Decision Lists DL
187
Support Vector Machines SVM
189
732 Empirical Evaluation on the DSO Corpus
190

19 Further Reading
21
References
22
Word Senses
29
22 Lexicographers
30
23 Philosophy
32
232 The Fregean Tradition and Reification
33
234 Implications for Word Senses
34
24 Lexicalization
35
25 Corpus Evidence
39
251 Lexicon Size
41
252 Quotations
42
26 Conclusion
43
27 Further Reading
44
Acknowledgments
45
Making Sense About Sense
47
32 WSD and the Lexicographers
49
33 WSD and Sense Inventories
51
34 NLP Applications and WSD
55
35 What Level of Sense Distinctions Do We Need for NLP If Any?
58
36 What Now for WSD?
64
37 Conclusion
68
Evaluation of WSD Systems
75
411 Terminology
76
412 Overview
80
42 Background
81
422 The Line and Interest Corpora
83
423 The DSO Corpus
84
424 Open Mind Word Expert
85
43 Evaluation Using PseudoWords
86
441 Senseval1
87
Evaluation and Scoring
88
English AllWords Task
89
443 Comparison of Tagging Exercises
91
45 Sources of InterAnnotator Disagreement
92
Groupings for WordNet
95
461 Criteria for WordNet Sense Grouping
96
462 Analysis of Sense Grouping
97
47 Senseval3
98
48 Discussion
99
References
102
KnowledgeBased Methods for WSD
107
52 Lesk Algorithm
108
521 Variations of the Lesk Algorithm
110
Simplified Lesk Algorithm
111
Augmented Semantic Spaces
113
53 Semantic Similarity
114
532 Using Semantic Similarity Within a Local Context
117
533 Using Semantic Similarity Within a Global Context
118
54 Selectional Preferences
119
Learning WordtoWord Relations
120
543 Using Selectional Preferences
122
55 Heuristics for Word Sense Disambiguation
123
552 One Sense Per Discourse
124
56 KnowledgeBased Methods at Senseval2
125
57 Conclusions
126
References
127
Unsupervised CorpusBased Methods for WSD
132
611 Scope
134
612 Motivation
136
Distributional Methods
137
Translational Equivalence
139
613 Approaches
140
62 TypeBased Discrimination
141
621 Representation of Context
142
622 Algorithms
145
Latent Semantic Analysis LSA
146
Hyperspace Analogue to Language HAL
147
Clustering By Committee CBC
148
623 Discussion
150
631 Representation of Context
151
Context Group Discrimination
152
McQuittys Similarity Analysis
154
633 Discussion
157
64 Translational Equivalence
158
641 Representation of Context
159
643 Discussion
160
65 Conclusions and the Way Forward
161
Acknowledgments
162
7 Supervised CorpusBased Methods for WSD
167
711 Machine Learning for Classification
168
An Example on WSD
170
72 A Survey of Supervised WSD
171
721 Main Corpora Used
172
722 Main Sense Repositories
173
723 Representation of Examples by Means of Features
174
724 Main Approaches to Supervised WSD
175
Methods Based on the Similarity of the Examples
176
Methods Based on Discriminating Rules
177
Methods Based on Rule Combination
179
Experiments
191
74 Current Challenges of the Supervised Approach
195
742 Porting Across Corpora
196
743 The Knowledge Acquisition Bottleneck
197
Automatic Acquisition of Training Examples
198
Active Learning
199
Parallel Corpora
200
744 Bootstrapping
201
745 Feature Selection and Parameter Optimization
202
746 Combination of Algorithms and Knowledge Sources
203
75 Conclusions and Future Trends
205
Acknowledgments
206
References
207
Knowledge Sources for WSD
217
82 Knowledge Sources Relevant to WSD
218
821 Syntactic
219
Collocations KS 3
220
Semantic Word Associations KS 6
221
Semantic Roles KS 8
222
Pragmatics KS 11
223
831 TargetWord Specific Features
224
832 Local Features
225
833 Global Features
227
84 Identifying Knowledge Sources in Actual Systems
228
841 Senseval2 Systems
229
842 Senseval3 Systems
231
851 Senseval Results
232
852 Yarowsky and Florian 2002
233
853 Lee and Ng 2002
234
854 Martínez et al 2002
237
855 Agirre and Martínez 2001a
238
856 Stevenson and Wilks 2001
240
86 Discussion
242
87 Conclusions
245
Acknowledgments
246
References
247
Automatic Acquisition of Lexical Information and Example
253
92 Mining Topical Knowledge About Word Senses
254
921 Topic Signatures
255
922 Association of Web Directories to Word Senses
257
93 Automatic Acquisition of SenseTagged Corpora
258
932 Bootstrapping from Seed Examples
261
933 Acquisition via Web Directories
263
934 Acquisition via CrossLanguage Evidence
264
935 WebBased Cooperative Annotation
268
94 Discussion
269
Acknowledgments
271
References
272
DomainSpecific WSD
275
102 Approaches to DomainSpecific WSD
277
1022 Topic Signatures and Topic Varietion
282
Topic Variation
283
1023 Domain Tuning
284
TopDown Domain Tuning
285
103 DomainSpecific Disambiguation in Applications
288
1032 CrossLingual Information Retrieval
289
1033 The MEANING Project
292
104 Conclusions
295
References
296
WSD in NLP Applications Philip Resnik University of Maryland
299
112 Why WSD?
300
Argument by Analogy
301
Argument from Specific Applications
302
113 Traditional WSD in Applications
303
1131 WSD in Traditional Information Retrieval
304
1132 WSD in Applications Related to Information Retrieval
307
Crosslanguage IR
308
Question Answering
309
Document Classification
312
1133 WSD in Traditional Machine Translation
313
1134 Sense Ambiguity in Statistical Machine Translation
315
1135 Other Emerging Applications
317
114 Alternative Conceptions of Word Sense
320
1142 Patterns of Usage
321
1143 CrossLanguage Relationships
323
115 Conclusions
325
References
326
A Resources for WSD
339
A12 Thesauri
341
A2 Corpora
343
A22 SenseTagged Corpora
345
A23 Automatically Tagged Corpora
347
A3 Other Resources
348
A32 Utilities Demos and Data
349
A33 Language Data Providers
350
Index of Terms
353
Index of Authors and Algorithms
361
Copyright

Other editions - View all

Common terms and phrases

Popular passages

Page ii - Bell Labs, New Jersey, USA Judith Klavans, Columbia University, New York, USA David T. Barnard, University ofRegina, Canada Dan Tufis, Romanian Academy of Sciences, Romania Joaquim Llisterri, Universitat Autonoma de Barcelona, Spain Stig Johansson, University of Oslo, Norway Joseph Mariani, LIMSI-CNRS, France...
Page 334 - Proceedings of the 19th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Zurich, 174-180.
Page xiv - School of Computing, National University of Singapore 3 Science Drive 2, Singapore...

References to this book

Bibliographic information