Synalp-Empathic: A Valence Shifting Hybrid System for Sentiment Analysis

Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 605–609, Dublin, Ireland, August 23-24, 2014. Synalp-Empathic: A Valence Shifting Hybrid System for Sentiment Analysis Alexandre Denis, Samuel Cruz-Lara, Nadia Bellalem and Lotfi Bellalem LORIA/University of Lorraine Nancy, France {alexandre.denis, samuel.cruz-lara, nadia.bellalem, lotfi.bellalem}@loria.fr Abstract This paper describes the Synalp-Empathic system that competed in SemEval-2014 Task 9B Sentiment Analysis in Twitter. Our system combines syntactic-based va- lence shifting rules with a supervised learning algorithm (Sequential Minimal Optimization). We present the system, its features and evaluate their impact. We show that both the valence shifting mech- anism and the supervised model enable to reach good results. 1 Introduction Sentiment Analysis (SA) is the determination of the polarity of a piece of text (positive, nega- tive, neutral). It is not an easy task, as proven by the moderate agreement between human an- notators when facing this task. Their agreement varies whether considering document or sentence level sentiment analysis, and different domains may show different agreements as well (Berming- ham and Smeaton, 2009). As difficult the task is for human beings, it is even more difficult for machines which face syn- tactic, semantic or pragmatic difficulties. Consider for instance irrealis phenomena such as “if this is good” or “it would be good if” that are both neu- tral. Irrealis is also present in questions (“is this good?”) but presupposition of existence does mat- ter: “can you fix this terrible printer?” would be polarized while “can you give me a good advice?” would not. Negation and irrealis interact as well, compare for instance “this could be good” (neutral or slightly positive) and “this could not be good” (clearly negative). Other difficult phenomena in- clude semantic or pragmatic effects, such as point This work is licensed under a Creative Commons At- tribution 4.0 International Licence. Page numbers and pro- ceedings footer are added by the organisers. Licence details: http://creativecommons.org/licenses/by/4.0/ of view (“Israel failed to defeat Hezbollah”, nega- tive for Israel, positive for Hezbollah), background knowledge (“this car uses a lot of gas”), seman- tic polysemy (“this vacuum cleaner sucks” vs “this movie sucks”), etc. From the start, machine learning has been the widely dominant approach to sentiment analy- sis since it tries to capture these phenomena all- together (Liu, 2012). Starting from simple n- grams (Pang et al., 2002), more recent approaches tend to include syntactic contexts (Socher et al., 2011). However these supervised approaches all require a training corpus. Unsupervised ap- proaches such as the seminal paper of (Turney, 2002) require training corpus as well but do not require annotations. We propose in this paper to look first at approaches that do not require any corpus because annotating a corpus is in general costly, especially in sentiment analysis in which several annotators are required to maintain a high level of agreement1. Nevertheless supervised ma- chine learning can be useful to adapt the system to a particular domain and we will consider it as well. Hence, we propose in this paper to first consider …

Content truncated to 500 words