Artificial Intelligence and Natural Language : 11th Conference, AINL 2022, Saint Petersburg, Russia, April 14–15, 2022, Revised Selected Papers

This book constitutes the refereed proceedings of the 11th Conference on Artificial Intelligence and Natural Language, AINL 2022, held in St. Petersburg, Russia, in April 2022. The 8 revised full papers and 1 short paper were carefully reviewed and selected from 20 submissions. The volume presents recent research in areas of of text mining, speech technologies, dialogue systems, information retrieval, machine learning, articial intelligence, and robotics. Preface Organization Contents Inferring Image Background from Text Description 1 Introduction 2 Related Work 3 Method 3.1 Searching Location Words in the Sentence 3.2 Searching Location Words in the Dictionary 4 Datasets 4.1 Real Texts 4.2 Synthetic Texts 5 Results 6 Conclusions References Topical Extractive Summarization 1 Introduction 2 Related Work 3 Model Description 3.1 ABAE 3.2 Heterogeneous LexRank 4 Datasets 4.1 Our New Dataset Constructed from WikiPersons 5 Experiments 5.1 Evaluation Metrics 5.2 Experiments with ABAE and Heterogeneous LexRank 6 Conclusion A Appendix A.1 Experiments with BigARTM and WikiPersons A.2 Experiments with Neural Topic Models and WikiPersons References The Semantic Shifts of the Topical Structure in the Corpus of Lentach News Posts 1 Introduction 2 Related Works 3 Experimental Design 3.1 Corpus Collecting and Preprocessing 3.2 BERT Topic Modeling 4 Results and Interpretation 5 Discussion 6 Summary References Development of Folklore Motif Classifier Using Limited Data 1 Introduction 2 Related Work 2.1 Zero-Shot Text Classification 2.2 Working with ATU Index Labeled Texts 3 Dataset 3.1 ATU Index 3.2 Texts 4 Approach 5 Experimental Setup 5.1 Vectorizing Texts 5.2 Models 6 Results 7 Conclusion References Morphological and Emotional Features of the Speech in Children with Typical Development, Autism Spectrum Disorders and Down Syndrome 1 Introduction 2 Datasets 3 Methods 3.1 Graphematic Analysis of the Texts 3.2 Morphological Text Analysis 4 Results 5 Conclusions and Discussion References WikiMulti: A Corpus for Cross-Lingual Summarization 1 Introduction 2 Existing Datasets 2.1 Multi-lingual Datasets 2.2 Cross-Lingual Datasets 3 WikiMulti Dataset 4 Experiments 4.1 Baselines 4.2 Experimental Setup 5 Results and Analysis 6 Conclusion A Evaluation results for the baseline models on WikiMulti B English Good articles divided into categories References Rethinking Crowd Sourcing for Semantic Similarity 1 Introduction 2 Related Work 3 Domain-Specific Annotator Attributes for Natural Language 4 The Data 5 Experiments 5.1 Filtering Heuristics for Unreliable Annotators 5.2 Correlation with Automated Semantic Similarity Metrics 6 Discussion 7 Conclusion References Interplay of Visual and Acoustic Cues of Irony Perception: A Case Study of Actor’s Speech 1 Introduction 2 Material 2.1 Linguistic Analysis 2.2 Requirements for the Audiovisual Material 3 Perceptual Experiments 4 Acoustic Analysis 5 Co-speech Gestures 6 Results and Discussion 7 Conclusions References Findings of Biomedical Russian to English Machine Translation Competition 1 Introduction 2 Dataset 2.1 Markup Process 3 Solutions 3.1 Baselines 3.2 Participant Solutions 4 Metrics 5 Competition Results 6 Conclusion References Translation of Medical Texts with Ensembling and Knowledge Distillation 1 Introduction 2 Related Work 3 Data Preparation 4 Description of the Approach 4.1 Byte-Pair Encoding 4.2 Single Model Description 4.3 Knowledge Distillation 4.4 Ensemble Teacher Model 5 Evaluation and Results 5.1 Single Teacher Model 5.2 Ensemble Teacher Model 6 Conclusion References Author Index