Enhanced Sentiment Learning Using Twitter Hashtags and Smileys

By Dmitry Davidov, Oren Tsur, Ari Rappoport. 2010

842 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Duyu Tang
2. Mohit Bansal
3. Percy Liang
4. Yejin Choi
5. Saif M. Mohammad
6. Richard Socher
7. Yoshua Bengio
8. Zhe Gan

Citations in the corpus

• "Blissfully Happy" or "Ready toFight": Varying Interpretations of Emoji (2016, methodology): “In research on using emoticons in sentiment analysis, Davidov, Tsur, and Rappoport (2010) found that when Amazon Mechanical Turk participants were presented with tweets in which emoticons had been removed, they were able to identify with high precision the original emoticon that had been in the…”
• Addressing Age-Related Bias in Sentiment Analysis (2018, background): “Natural language processing techniques, including sentiment analysis, are a primary point for inquiry among both social scientists and machine learning researchers [11,18,33,51].”
• Predictive Analysis on Twitter: Techniques and Applications (2018, methodology): “Also, text processing techniques like stopwords removal, word-pattern identification, and punctuation removal have shown to improve sentiment analysis in [17].”
• Global Belief Recursive Neural Networks (2014, background): “Sentiment analysis has been the subject of research for some time [4, 2, 3, 6, 1, 23].”
• Robust Image Sentiment Analysis Using Progressively Trained and Domain Transferred Deep Networks (2015, background): “The details of the proposed framework will be described in the following sections.”
• SemEval-2019 Task 3: EmoContext Contextual Emotion Detection in Text (2019, methodology): “…of sentences which are 3 Task webpage: humanizing-ai.com/emocontext.html then used as input by classifiers such as Decision Trees, SVMs among others to predict the output (Alm et al., 2005), (Balabantaray et al., 2012), (Davidov et al., 2010), (Kunneman et al., 2014), (Yan and Turtle, 2016).”
• Sentiment of Emojis (2015, methodology): “A small set of emoticons has already been used as additional features for polarity classification [8, 20].”
• #TagSpace: Semantic Embeddings from Hashtags (2014, background): “Some previous work (Davidov et al., 2010; Godin et al., 2013; She and Chen, 2014) has addressed hashtag prediction. Most such work applies to much smaller sets of hashtags than the 100,000 we consider, with the notable exception of Ding et al. (2012), which uses an unsupervised method.”
• Twitter Polarity Classification with Label Propagation over Lexical Links and the Follower Graph (2011, methodology): “by taking emoticons like ‘:)’ as positive and ‘:(’ as negative, and train a standard classifier (Read, 2005; Go et al., 2009).1 Semi-supervised methods can also reduce dependence on labeled texts: for example, Sindhwani and Melville (2008) use a polarity lexicon combined with label propagation.”
• Detecting offensive tweets via topical feature discovery over a large scale twitter corpus (2012, background): “Twitter has drawn significant attention in recent years, and much work related to sentiment analysis has been published, such as [5, 4, 9].”
• Sentiment Analysis and Opinion Mining (2012, background): “In Davidov et al. (2010), sentiment classification of Twitter postings (or tweets) was studied.”
• Target-dependent Twitter Sentiment Classification (2011, background): “(Go et al., 2009; Parikh and Movassate, 2009; Barbosa and Feng, 2010; Davidiv et al., 2010) all follow the machine learning based approach for sentiment classification of tweets.”
• #Communing affiliation: Social tagging as a resource for aligning around values in social media (2017, background): “Discourse Context Media (2017), http://dx.doi.org/10.1016/j.dcm.2017.08.001 using tags to improve natural language processing tasks such as identifying and categorising sentiment (Davidov et al., 2010; Kunneman et al., 2015; Mohammad and Kiritchenko, 2015).”
• Learning Sentiment-Specific Word Embedding for Twitter Sentiment Classification (2014, background): “Many studies on Twitter sentiment classification (Pak and Paroubek, 2010; Davidov et al., 2010; Barbosa and Feng, 2010; Kouloumpis et al., 2011; Zhao et al., 2012) leverage massive noisy-labeled tweets selected by positive and negative emoticons as training set and build sentiment classifiers…”
• VADER: A Parsimonious Rule-Based Model for Sentiment Analysis of Social Media Text (2014, result): “Consistent with prior work (Agarwal, Xie, Vovsha, Rambow , & Passonneau, 2011; Davidov et al., 2010; Shastri, Parvathy, Kumar, Wesley, & Balakrishnan, 2010), we find that grammatical features (conventions of use for punctuation and capitalization) and consideration for degree modifiers like " very…”