Intro -- Preface -- Organization -- Contents -- PolSentiLex: Sentiment Detection in Socio-Political Discussions on Russian Social Media -- 1 Introduction -- 2 Related Work -- 2.1 Sentiment Analysis in the Russian Langauge -- 3 PolSentiLex -- 3.1 LiveJournal Collection of Social and Political Posts -- 3.2 Selection of Potentially Sentiment-Bearing Words -- 3.3 Data Mark Up -- 3.4 The Three Versions of PolSentiLex -- 4 PolSentiLex Quality Assessment -- 4.1 Datasets -- 5 Results -- 6 Conclusion -- References
Automatic Detection of Hidden Communities in the Texts of Russian Social Network Corpus -- Abstract -- 1 Introduction -- 2 Related Works -- 3 Experiments with the Russian Corpus of VKontakte Posts -- 3.1 Corpus Collecting and Preprocessing -- 3.2 Author-Topic Models -- 3.3 Automatic Labeling of Topics -- 3.4 Model of Hidden Communities in VKontakte Social Network -- 4 Results and Evaluation -- 5 Summary -- References -- Dialog Modelling Experiments with Finnish One-to-One Chat Data -- 1 Introduction -- 2 Related Work -- 2.1 Language Resources for One-to-One Chat Dialogue Data
2.2 Machine Learning for Chat Data Modelling -- 2.3 Evaluation Results for Chat Data Models -- 3 Experimental Setting -- 3.1 Description of the Data Sets -- 3.2 Implementation and Parameters of Methods -- 3.3 Data Preprocessing -- 4 Results -- 4.1 Output Examples -- 5 Analysis and Discussion -- 6 Conclusions -- References -- Advances of Transformer-Based Models for News Headline Generation -- 1 Introduction -- 2 Related Work -- 3 Models Description -- 4 Datasets -- 5 Experiments -- 5.1 Evaluation -- 5.2 Training Dynamics -- 6 Results -- 6.1 Human Evaluation -- 6.2 Error Analysis
7 Conclusion and Future Work -- References -- An Explanation Method for Black-Box Machine Learning Survival Models Using the Chebyshev Distance -- 1 Introduction -- 2 Basic Definitions of Survival Analysis -- 3 LIME -- 4 A General Algorithm of SurvLIME and SurvLIME-Inf -- 5 Optimization Problem for Computing Parameters -- 6 Numerical Experiments -- 6.1 Synthetic Data -- 6.2 Real Data -- 7 Conclusion -- References -- Unsupervised Neural Aspect Extraction with Related Terms -- 1 Introduction -- 2 Related Work -- 3 The Proposal -- 3.1 Model -- 3.2 Training Objective -- 4 Experiments -- 4.1 Datasets
4.2 Experimental Settings -- 4.3 Evaluation Settings -- 4.4 Aspect Extraction Results -- 4.5 Aspect and Aspect Term Extraction Results -- 5 Conclusions -- References -- Predicting Eurovision Song Contest Results Using Sentiment Analysis -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Collection of Eurovision Tweets -- 3.2 Identification of the Source Country -- 3.3 Tweet Tokenization -- 3.4 Identification of the Target Country -- 3.5 Sentiment Analysis -- 3.6 Tallying of Final Results -- 4 Experimental Results -- 4.1 Televoting Algorithm -- 4.2 Different Sampling Windows
Notes
International conference proceedings
"Originally planned to take place at Helsinki in Finland, AINL 2020 was held as a fully digital conference during October 7-9."
Includes author index
Online resource; title from PDF title page (SpringerLink, viewed November 24, 2020)