Anaphora Resolution and Text Retrieval: A Linguistic Analysis of Hypertexts

Description

VFI Foundation Award 2013 This book covers anaphora resolution for the English language from a linguistic and computational point of view. First, a definition of anaphors that applies to linguistics as well as information technology is given. On this foundation, all types of anaphors and their characteristics for English are outlined. To examine how frequent each type of anaphor is, a corpus of different hypertexts has been established and analysed with regard to anaphors. The most frequent type are non-finite clause anaphors - a type which has never been investigated so far. Therefore, the potential of non-finite clause anaphors are further explored with respect to anaphora resolution. After presenting the fundamentals of computational anaphora resolution and its application in text retrieval, rules for resolving non-finite clause anaphors are established. Therefore, this book shows that a truly interdisciplinary approach can achieve results which would not have been possible otherwise. Open Access: In July 2019, this volume was retroactively turned into an Open Access publication thanks to the support of the Fachinformationsdienst Linguistik. https://www.linguistik.de/ Foreword Contents 1 Introduction 2 Linguistic fundamentals of anaphors and anaphora 2.1 Basic definitions 2.2 The relationship between anaphor and antecedent 2.2.1 Coreference 2.2.2 Substitution 2.2.3 Comparison of coreference and substitution 2.2.4 Sense relations 2.3 Related concepts: cataphora, endophora, exophora and deixis 2.3.1 Cataphora 2.3.2 Endophora and exophora 2.3.3 Deixis 2.4 Anaphors as cohesive devices in texts 2.4.1 Texts and their features 2.4.2 Cohesion 2.4.3 Cohesive devices as a form of reduction 2.5 Anaphors in the present book 3 Types of anaphors 3.1 Central pronouns 3.1.1 Personal pronouns 3.1.1.1 Subjective and objective forms 3.1.1.2 Person, number and gender 3.1.1.3 Anaphoric and non-anaphoric use 3.1.1.4 Cataphoric use 3.1.1.5 Relationship between anaphor and antecedent 3.1.1.6 Summary 3.1.2 Possessive pronouns 3.1.2.1 Determinative and independent possessive pronouns 3.1.2.2 Person, number and gender 3.1.2.3 Anaphoric/cataphoric and non-anaphoric use 3.1.2.4 Relationship between anaphor and antecedent 3.1.2.5 Summary 3.1.3 Reflexive pronouns 3.1.3.1 Basic and emphatic use 3.1.3.2 Person, number and gender 3.1.3.3 Anaphoric/cataphoric and non-anaphoric use 3.1.3.4 Summary 3.1.4 Summary of personal, possessive and reflexive pronouns 3.2 Reciprocal pronouns 3.2.1 Compound and split construction 3.2.2 Anaphoric/cataphoric and non-anaphoric use 3.2.3 Summary 3.3 Demonstrative pronouns 3.3.1 Dependent and independent demonstrative pronouns 3.3.2 Grammatical and referential aspects 3.3.3 Anaphoric/cataphoric and non-anaphoric use 3.3.4 Relationship between anaphor and antecedent 3.3.5 Summary 3.4 Relative pronouns 3.4.1 Form and function 3.4.2 Types of clauses and their anaphoric and non-anaphoric use 3.4.3 Further non-anaphoric uses 3.4.4 Gender and case 3.4.5 Summary 3.5 Adverbs 3.5.1 Here and there 3.5.2 Now and then 3.5.3 Where, when, while and why 3.5.4 Summary 3.6 Noun phrases with a definite article 3.6.1 Syntactic characteristics 3.6.2 Anaphoric/cataphoric and non-anaphoric use 3.6.3 The relationship between anaphor and antecedent 3.6.4 Summary 3.7 Proper names 3.7.1 Proper names and proper nouns 3.7.2 Syntactic features 3.7.3 Anaphoric and non-anaphoric use 3.7.4 Summary 3.8 Indefinite pronouns 3.8.1 One and ones 3.8.2 Other, others and another 3.8.3 Both, all and each 3.8.4 Enough and several 3.8.5 Some and any 3.8.6 Either, neither and none 3.8.7 Many and much/more/most, few/fewer/fewest and little/less/least 3.8.8 Further non-anaphoric uses of all of-pronouns 3.8.9 Cataphoric use 3.8.10 Indefinite pronouns as elliptical forms 3.8.11 Summary 3.9 Other forms of coreference and substitution: the same, such and so 3.9.1 The same 3.9.2 Such 3.9.3 So 3.9.4 Cataphoric use 3.9.5 Summary 3.10 Verb phrases with do and combinations with so, this, that, it and the same (thing) 3.10.1 Simple forms with do 3.10.2 Complex forms with do 3.10.3 The form of the antecedent and cataphoric use 3.10.4 The relationship between anaphor and antecedent 3.10.5 Summary 3.11 Ellipses 3.11.1 Types of ellipses 3.11.2 Non-anaphoric ellipsis 3.11.3 Summary 3.12 Non-finite clauses 3.12.1 To-infinitive clauses 3.12.1.1 Anaphoric/cataphoric and non-anaphoric use with regard to clause and phrase functions 3.12.1.2 Further non-anaphoric uses 3.12.1.3 Summary 3.12.2 -ing-participle clauses 3.12.2.1 Anaphoric/cataphoric and non-anaphoric use with regard to clause and phrase functions 3.12.2.2 Further non-anaphoric uses 3.12.2.3 Summary 3.12.3 -ed-participle clauses 3.12.3.1 Anaphoric/cataphoric and non-anaphoric use with regard to clause and phrase functions 3.12.3.2 Further non-anaphoric uses 3.12.3.3 Summary 3.12.4 Summary of to-infinitive, -ing-participle and -ed-participle clauses 3.13 Conclusion 4 Anaphors in hypertexts 4.1 Previous research 4.2 Hypertexts 4.2.1 What are hypertexts? 4.2.2 Cohesion and coherence in hypertexts 4.2.3 Linguistic characteristics of hypertexts 4.2.4 Classifications of hypertexts 4.3 Corpus of hypertexts 4.3.1 Corpus design 4.3.2 “Wikipedia” 4.3.3 “Blogs” 4.3.4 “(Traditional) Websites” 4.4 Results of the corpus analysis 4.4.1 Frequency of anaphors in the corpus 4.4.1.1 Distribution of the twelve anaphor types 4.4.1.2 Central pronouns 4.4.1.3 Reciprocal pronouns 4.4.1.4 Demonstrative pronouns 4.4.1.5 Relative pronouns 4.4.1.6 Adverbs 4.4.1.7 Proper names 4.4.1.8 Indefinite pronouns 4.4.1.9 Other forms of coreference and substitution 4.4.1.10 Verb phrases with do and combinations 4.4.1.11 Ellipses 4.4.1.12 Non-finite clauses 4.4.2 Ratio of items occurring anaphorically and non-anaphorically 4.5 Conclusion 5 Text retrieval and its handling of anaphors 5.1 What are text retrieval systems? 5.2 Text retrieval models 5.3 Evaluation of text retrieval results 5.4 Natural language processing methods in text retrieval 5.4.1 Sentence delimitation and tokenisation 5.4.2 Stop word detection 5.4.3 Stemming and lemmatisation 5.4.4 Part-of-speech tagging 5.4.5 Parsing 5.4.6 Natural language processing on the Web 5.5 Text retrieval and the Internet 5.5.1 Crawling 5.5.2 Indexing 5.5.2.1 Processes of feature extraction 5.5.2.2 Example of a feature extraction 5.5.2.3 Index 5.5.3 Queries and search results 5.6 Conclusion 6 Approaches to and uses of anaphora resolution 6.1 Uses of anaphora resolution systems 6.1.1 Machine translation 6.1.2 Information extraction 6.1.3 Question answering 6.1.4 Text summarisation 6.1.5 Text retrieval 6.2 Structure of anaphora resolution systems 6.3 Anaphora resolution approaches 6.3.1 Rule-based approaches 6.3.2 Data-based approaches 6.3.3 Comparison of anaphora resolution approaches 6.4 Anaphora resolution in text retrieval 6.5 Evaluation of anaphora resolution systems 6.5.1 Measures for evaluation 6.5.2 Comparing different algorithms 6.5.3 Annotated corpora for evaluation 6.5.3.1 Current annotated corpora 6.5.3.2 An annotated corpus for all anaphor types 6.6 Conclusion 7 Development of extensive linguistic rules for anaphora resolution: the example of non-finite clause anaphors 7.1 Identifying anaphoric items 7.1.1 Rules for -ing-items and their evaluation 7.1.2 Rules for -ed-items and their evaluation 7.1.3 Rules for to-items and their evaluation 7.2 Identifying antecedents 7.2.1 Rules for -ing and their evaluation 7.2.2 Rules for -ed and their evaluation 7.2.3 Rules for to and their evaluation 7.3 Conclusion 8 Conclusion Bibliography Index