Formalizing natural languages - applications to natural language processing and digital humanities : 16th International Conference, NOOJ 2022, Rosario, Argentina, June 14-16, 2022, revised selected papers

This book constitutes selected revised papers of the 16th International Conference on Formalizing Natural Languages: Applications to Natural Language Processing and Digital Humanities, NooJ 2022, held in Rosario, Argentina, in June 2022. Due to COVID-19 pandemic the conference was held virtually. NooJ is a linguistic development environment that provides tools for linguists to construct linguistic resources that formalize a large gamut of linguistic phenomena: typography, orthography, lexicons for simple words, multiword units and discontinuous expressions, inflectional, derivational and agglutinative morphology, local, phrase-structure and dependency grammars, as well as transformational and semantic grammars. The 17 full papers presented were carefully reviewed and selected from 50 submissions. The papers are organized in the following topics:​ Morphological and Lexical Resources; Syntactic and Semantic Resources; Corpus Linguistics and Discourse Analysis; Natural Language Processing Applications. Preface 6 Organization 11 Contents 13 List of Contributors 15 Morphological and Lexical Resources 17 The Architecture of SANTI-Morf’s Guesser Module 18 1 To Guess or Not to Guess 18 2 A Brief Look at SANTI-Morph’s Architecture 19 3 The Guesser 20 3.1 Morphological Cues 20 3.2 Capitalization 21 3.3 Final Guess 22 3.4 Ambiguity Problem 22 3.5 Solution 23 3.6 Architecture 24 4 The New +UNAMB Operator: A Proposal for Simplification 26 5 Conclusion 27 References 27 Formation and Evolution of Intensive Adverbs Ending in -mente Derived from the Adjectival Class in Spanish and French 29 1 Introduction 29 2 Intensive Adjectives Derived from the Class 31 2.1 Terriblemente/Terriblement 31 2.2 Horriblemente/Horriblement 33 2.3 Espantosamente/Épouvantablement 34 2.4 Other Adverbs 35 3 Conclusion 38 References 39 Formalizing the Ancient Greek Participle Inflection with NooJ 41 1 Introduction 41 1.1 The Ancient Greek Participle 41 1.2 The Modern Greek Module 42 1.3 The Text Corpus and the Participle Corpus 42 2 Morphological Processing of the Ancient Greek Participle 43 2.1 Extraction of Participles 43 2.2 Selection of Participles 44 2.3 Lemmas Classification Criteria: Accented Vowels Before Verb Endings 45 2.4 Creation of Dictionaries and Grammars 46 2.5 Linguistic Analysis and TAS 50 3 Conclusions, Challenges and Perspectives 53 References 53 Automatic Analysis of Appreciative Morphology: The Case of Paronomasia in Colombian Spanish 54 1 Introduction 54 2 Theoretical Framework: The Phenomenon of Paronomasia, Generative Lexicon Theory and Constraint Grammar 55 2.1 Paronomasia: Description of the Phenomenon 55 2.2 The Generative Lexicon Theory 57 2.3 Constraint Grammars 57 3 Methodology 58 3.1 Selection of Colombian Spanish Names 58 3.2 Computer Work 59 4 Results Obtained 62 5 Conclusions 64 References 64 Prosodic Segmentation of Belarusian Texts in NooJ 65 1 Introduction 65 2 Relevance of the Study: Lack of Automatic Prosodic Segmentation in TTS 66 3 Text Sources: Corpus of Literary and Medical Texts 67 4 Extraction of Syntagms 68 5 Syntactic Grammar for Extracting Syntagms in NooJ 72 6 Conclusion 76 References 76 Syntactic and Semantic Resources 78 Zellig S. Harris’ Transfer Grammar and Its Application with NooJ 79 1 Introduction 79 2 On-Line Translation Portals and Tools 80 2.1 A Brief Outline of Google Translate Errors 80 3 Zellig S. Harris’ Transfer Grammar 83 4 NooJ and Harris’ Transfer Grammar 85 5 Conclusions and Future Work 89 References 89 Formalization of Transformations of Complex Sentences in Quechua 90 1 Introduction 90 2 Gerund Subordination Classes 91 2.1 Concordance Grammars 93 3 Basic Transformations (BT) 94 3.1 Permutation [PERM] 94 3.2 Interrogation [INT] 95 3.3 Past2 [PASSG] 95 3.4 Past1 [PASSG1] 96 3.5 Future [F] 97 3.6 Progressive [Prog2] 98 4 Paraphrasing Subordinate Clauses 99 5 Transformation and Machine Translation 100 6 Conclusion 101 References 101 Automatic Extraction of Verbal Phrasemes in the Electrical Energy Field with NooJ 103 1 Introduction 103 2 Phraseology Criteria 104 2.1 Frequency 105 2.2 Fixation 105 2.3 Pragmatic Criterion 105 3 Language Description Software: NOOJ’S Approach 106 3.1 Rational Grammar 106 3.2 Algebraic Grammar 107 3.3 Contextual Grammar 107 4 Modeling and Implementation in NooJ 109 5 Results 110 6 Conclusion 111 References 111 A Linguistic Approach for Automatic Analysis, Recognition and Translation of Arabic Nominal Predicates 114 Abstract 114 1 Introduction 114 2 Nominal Predicates, Support Verbs and Translation 115 3 Linguistic Characteristics of the Nominative Predicate: Advice 115 3.1 The General Support Verbs: أفعال ناقلة عامّة 116 3.2 The Aspectual Support Verbs: أفعال ناقلة مظهريّة 116 3.3 Determiners of Predicate “Advice” 117 3.4 The Verbal Predicate “نَصَحَ” “Conseiller” “ to Advise” 117 4 Creation of a Bilingual French-Arabic Dictionary 117 5 The Creation of Grammars for the Analysis and Recognition of Syntactic Patterns 120 6 The Creation of Translation Grammars 121 7 Conclusion and Perspectives 124 References 125 Corpus Linguistics and Discourse Analysis 126 Processing the Discourse of Insecurity in Rosario with the NooJ Platform 127 1 Introduction 127 1.1 The Aim of Our Paper 127 1.2 Why Do We Use NooJ? 128 1.3 The Corpus 128 2 General Considerations 129 2.1 New Tags 129 2.2 Grammars for the Discourse of Insecurity 131 3 Conclusions 136 References 137 Analyzing Political Discourse: Finding the Frames for Guilt and Responsibility 139 1 Introduction 139 2 Previous Work 140 3 About the Language and the Corpus 141 3.1 Croatian Language Specifics 141 3.2 Croatian Parliament Corpus 142 4 Design of the Language Models 143 4.1 Language Model for ‘Responsibility’ 144 4.2 Language Model for ‘Guilt’ 146 5 Discussion of Results 147 6 Conclusion 149 References 150 Creation of Parallel Medical and Social Domains Corpora for the Machine Translation and Speech Synthesis Systems 151 Abstract 151 1 First Section Introduction 151 2 Compilation of a Medical and Social Domain Corpus 152 3 Creating Medical and Social Trilingual Dictionaries in NooJ 155 4 NooJ for Proofreading and Translating 158 4.1 Common Names in the Corpora 158 4.2 Personal Names in the Corpora 159 5 Conclusion 161 References 161 Creation of a Legal Domain Corpus for the Belarusian Module in NooJ: Texts, Dictionaries, Grammars 163 Abstract 163 1 Introduction 163 2 Compilation of a Legal Domain Corpus 164 3 Creating Medical and Social Trilingual Dictionaries in NooJ 168 4 Syntactic Grammars for Searching Legal Terminology and Punctuational Syntagmas 171 5 Conclusion 173 References 174 Natural Language Processing Applications 175 Construction of an Educational Game “CONJNOOJ” 176 1 Introduction 176 2 Proposed Method 177 2.1 Identification of Resources 178 2.2 Building of Resources 178 2.3 Game Steps 183 3 CONJNooJ Experimentation 183 3.1 Conjugation of Verbs 183 3.2 Inflection of Adjectives and Nouns 186 4 Conclusion and Perspectives 188 References 188 Annotation of Procedural Questions in Standard Arabic Using Syntactic Grammars 189 Abstract 189 1 Introduction 189 2 Related Works 190 3 The Generic Architecture of the Proposed QA System 191 4 Our Approach 192 4.1 Question Analysis 192 4.2 Segmentation and Passage Retrieval 194 5 Experiments and Results 197 5.1 Named Entity Recognition (NER) 197 5.2 Recognition and Analysis of Procedural Questions 198 6 Conclusion 198 References 199 Integrated NooJ Environment for Arabic Linguistic Disambiguation Improvement Using MWEs 200 Abstract 200 1 Introduction 200 2 State of Art 201 2.1 Arabic Specificities 201 2.2 Sources of Syntactic Ambiguity in Arabic 202 2.2.1 Vocalization Ambiguity 202 2.2.2 Word Order Ambiguity 202 2.2.3 Multifunctionality of Arabic Nouns 203 2.3 Arabic Disambiguation Task 203 3 Approach 204 3.1 Disambiguation Approach 204 3.2 Disambiguation Grammars 205 4 Experimentations and Results 208 5 Conclusion 210 Appendix 211 References 211 The Digital Text Workshop Cloud, New Solutions for Super Calculation Environments 213 1 Introduction 214 2 Comparing Models: Language Environments and Lexicon-Grammar as an Elementary Calculation LGLI 217 3 The Lexicon-Grammatical and Automatic Text Analysis 220 4 Digital Intelligence W.Tool in NooJ Environment 222 5 Conclusion and Research Results 223 References 223 Author Index 225