Target Two Birds With One SToNe: Entity-Level Sentiment and Tone Analysis in Croatian News Headlines

Sentiment analysis is often used to examine how different actors are portrayed in the media, and analysis of news headlines is of particular interest due to their attention-grabbing role. We address the task of entity-level sentiment analysis from Croatian news headlines. We frame the task as targeted sentiment analysis (TSA), explicitly differentiating between sentiment toward a named entity and the overall tone of the headline. We describe SToNe, a new dataset for this task with sentiment and tone labels. We implement several neural benchmark models, utilizing single- and multi-task training, and show that TSA can benefit from tone information. Finally, we gauge the difficulty of this task by leveraging dataset cartography.

[1]  J. Šnajder,et al.  ALANNO: An Active Learning Annotation System for Mortals , 2022, EACL.

[2]  W. Yu,et al.  A Survey of Multi-task Learning in Natural Language Processing: Regarding Task Relatedness and Training Methods , 2022, EACL.

[3]  Damián A. Furman,et al.  A Spanish dataset for Targeted Sentiment Analysis of political headlines , 2022, ArXiv.

[4]  Arzucan Özgür,et al.  A Dataset and BERT-based Models for Targeted Sentiment Analysis on Turkish Texts , 2022, ACL.

[5]  Bela Gipp,et al.  Towards Target-Dependent Sentiment Classification in News Articles , 2021, iConference.

[6]  N. Ljubešić,et al.  BERTić - The Transformer Language Model for Bosnian, Croatian, Montenegrin and Serbian , 2021, BSNLP.

[7]  Marcin Sydow,et al.  A dataset for Sentiment analysis of Entities in News headlines (SEN) , 2021, KES.

[8]  Yejin Choi,et al.  Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics , 2020, EMNLP.

[9]  Blaž Škrlj,et al.  Zero-Shot Learning for Cross-Lingual News Sentiment Classification , 2020, Applied Sciences.

[10]  Quoc V. Le,et al.  ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators , 2020, ICLR.

[11]  Joshua M. Scacco,et al.  The curiosity effect: Information seeking in the contemporary news environment , 2019, New Media Soc..

[12]  Sebastian Stabinger,et al.  Adapt or Get Left Behind: Domain Adaptation through BERT Language Model Finetuning for Aspect-Target Sentiment Classification , 2019, LREC.

[13]  Theo Araujo,et al.  What’s the Tone? Easy Doesn’t Do It: Analyzing Performance and Agreement Between Off-the-Shelf Sentiment Analysis Tools , 2019, Communication Methods and Measures.

[14]  Ao Feng,et al.  Target-Dependent Sentiment Classification With BERT , 2019, IEEE Access.

[15]  Zhen Huang,et al.  Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification , 2019, ACL.

[16]  Jiaxin Pei,et al.  Targeted Sentiment Analysis: A Data-Driven Categorization , 2019, ArXiv.

[17]  Tao Jiang,et al.  Attentional Encoder Network for Targeted Sentiment Classification , 2019, ICANN.

[18]  Xin Li,et al.  A Unified Model for Opinion Target Extraction and Target Sentiment Prediction , 2018, AAAI.

[19]  Nikola Ljubešić,et al.  What does Neural Bring? Analysing Improvements in Morphosyntactic Annotation and Lemmatisation of Slovenian, Croatian and Serbian , 2019, BSNLP@ACL.

[20]  Charlotte Galpin,et al.  Converging towards Euroscepticism? Negativity in news coverage during the 2014 European Parliament elections in Germany and the UK , 2018, European Politics and Society.

[21]  Sashank J. Reddi,et al.  On the Convergence of Adam and Beyond , 2018, ICLR.

[22]  Anne Schuth,et al.  Effective Headlines of Newspaper Articles in a Digital Environment , 2017 .

[23]  Nikola Ljubesic,et al.  New Inflectional Lexicons and Training Corpora for Improved Morphosyntactic Annotation of Croatian and Serbian , 2016, LREC.

[24]  Bosko Nikolic,et al.  Reliable Baselines for Sentiment Analysis in Resource-Limited Languages: The Serbian Movie Review Dataset , 2016, LREC.

[25]  Navneet Kaur,et al.  Opinion mining and sentiment analysis , 2016, 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom).

[26]  Yue Zhang,et al.  Gated Neural Networks for Targeted Sentiment Analysis , 2016, AAAI.

[27]  Stuart Soroka,et al.  Bad News or Mad News? Sentiment Scoring of Negativity, Fear, and Anger in News Content , 2015 .

[28]  M. Grbeša.  Framing of the President: Newspaper Coverage of Milan Bandić and Ivo Josipović in the Presidential Elections in Croatia in 2010 , 2013 .

[29]  Frank Esser,et al.  Negativity in political news: A review of concepts, operationalizations and key findings , 2012 .

[30]  Claes H. de Vreese,et al.  Effects of Election News Coverage: How Visibility and Tone Influence Party Choice , 2010 .

[31]  C. D. Vreese,et al.  News framing: Theory and typology , 2005 .

[32]  Daniel Dor,et al.  On newspaper headlines as relevance optimizers , 2003 .

[33]  Jr. James W.Tankard.  The Empirical Approach to the Study of Media Framing , 2001 .

[34]  P. Valkenburg,et al.  Framing European politics: a content analysis of press and television news , 2000 .

[35]  J. R. Landis,et al.  The measurement of observer agreement for categorical data. , 1977, Biometrics.