NewsMTSC: A Dataset for (Multi-)Target-dependent Sentiment Classification in Political News Articles

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 1663–1675 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 1663 NewsMTSC: A Dataset for (Multi-)Target-dependent Sentiment Classification in Political News Articles Felix Hamborg1,2 Karsten Donnay3,2 1Dept. of Computer and Information Science, University of Konstanz, Germany 2Heidelberg Academy of Sciences and Humanities, Germany 3Dept. of Political Science, University of Zurich, Switzerland felix.hamborg@uni-konstanz.de donnay@ipz.uzh.ch Abstract Previous research on target-dependent senti- ment classification (TSC) has mostly focused on reviews, social media, and other domains where authors tend to express sentiment explic- itly. In this paper, we investigate TSC in news articles, a much less researched TSC domain despite the importance of news as an essential information source in individual and societal decision making. We introduce NewsMTSC, a high-quality dataset for TSC on news articles with key differences compared to established TSC datasets, including, for example, differ- ent means to express sentiment, longer texts, and a second test-set to measure the influence of multi-target sentences. We also propose a model that uses a BiGRU to interact with multiple embeddings, e.g., from a language model and external knowledge sources. The proposed model improves the performance of the prior state-of-the-art from F1m = 81.7 to 83.1 (real-world sentiment distribution) and from F1m = 81.2 to 82.5 (multi-target sen- tences). 1 Introduction Previous research on target-dependent sentiment classification (TSC, also called aspect-term senti- ment classification) and related tasks, e.g., aspect- based sentiment classification (ABSC, or aspect- category sentiment classification) and stance de- tection, has focused mostly on domains in which authors tend to express their opinions explicitly, such as reviews and social media (Pontiki et al., 2015; Nakov et al., 2016; Rosenthal et al., 2017; Zhang et al., 2020; Hosseinia et al., 2020; AlDayel and Magdy, 2020; Liu, 2012; Zhang et al., 2018). We investigate TSC in news articles – a much less researched domain despite its critical relevance, especially in times of “fake news,” echo chambers, and news ownership centralization (Hamborg et al., 2019). How persons are portrayed in news on polit- ical topics is very relevant, e.g., for individual and societal opinion formation (Bernhardt et al., 2008). We define our problem statement as follows: we seek to detect polar judgments towards target per- sons (Steinberger et al., 2017). Following the TSC literature, we include only in-text, specifically in- sentence, means to express sentiment. In news texts such means are, e.g., word choice and generally framing1 (Kahneman and Tversky, 1984; Entman, 2007), e.g., “freedom fighters” vs. “terrorists,” or describing actions performed by the target, and in- direct sentiment through quoting another person (Steinberger et al., 2017). Other means may also al- ter the perception of persons and topics in the news, but are not in the scope of the task (Balahur et al., 2010), e.g., because they are not on sentence-level. For example, story selection, source selection, ar- ticle’s placement and size (Hamborg et al., 2019), and epistemological bias (Recasens et al., 2013). The main contributions of this …

Content truncated to 500 words