In recent years, the amount of user-generated opinionated texts (e.g., reviews, user comments) continues to grow at a rapid speed: featured news stories on a major event easily attract thousands of user comments on a popular online News service. How to consume subjective information of this volume becomes an interesting and important research question. In contrast to previous work on review analysis that tried to filter or summarize information for a generic average user, we explore a different direction of enabling personalized recommendation of such information. For each user, our task is to rank the comments associated with a given article according to personalized user preference (i.e., whether the user is likely to like or dislike the comment). To this end, we propose a factor model that incorporates rater-comment and rater-author interactions simultaneously in a principled way. Our full model significantly outperforms strong baselines as well as related models that have been considered in previous work. 1

This paper explores approaches to sentiment classification of U.S. Congressional floordebate transcripts. Collective classification techniques are used to take advantage of the informal citation structure present in the debates. We use a range of methods based on local and global formulations and introduce novel approaches for incorporating the outputs of machine learners into collective classification algorithms. Our experimental evaluation shows that the mean-field algorithm obtains the best results for the task, significantly outperforming the benchmark technique. 1

Aspect extraction is a central problem in sentiment analysis. Current methods either extract aspects without categorizing them, or extract and categorize them using unsupervised topic modeling. By categorizing, we mean the synonymous aspects should be clustered into the same category. In this paper, we solve the problem in a different setting where the user provides some seed words for a few aspect categories and the model extracts and clusters aspect terms into categories simultaneously. This setting is important because categorizing aspects is a subjective task. For different application purposes, different categorizations may be needed. Some form of user guidance is desired. In this paper, we propose two statistical models to solve this seeded problem, which aim to discover exactly what the user wants. Our experimental results show that the two proposed models are indeed able to perform the task effectively. 1

This paper presents a survey on the role of negation in sentiment analysis. Negation is a very common linguistic construction that affects polarity and, therefore, needs to be taken into consideration in sentiment analysis. We will present various computational approaches modeling negation in sentiment analysis. We will, in particular, focus on aspects, such as level of representation used for sentiment analysis, negation word detection and scope of negation. We will also discuss limits and challenges of negation modeling on that task. 1

A growing body of work has highlighted the challenges of identifying the stance a speaker holds towards a particular topic, a task that involves identifying a holistic subjective disposition. We examine stance classification on a corpus of 4873 posts across 14 topics on ConvinceMe.net, ranging from the playful to the ideological. We show that ideological debates feature a greater share of rebuttal posts, and that rebuttal posts are significantly harder to classify for stance, for both humans and trained classifiers. We also demonstrate that the number of subjective expressions varies across debates, a fact correlated with the performance of systems sensitive to sentimentbearing terms. We present results for identifing rebuttals with 63 % accuracy, and for identifying stance on a per topic basis that range from 54 % to 69%, as compared to unigram baselines that vary between 49 % and 60%. Our results suggest that methods that take into account the dialogic context of such posts might be fruitful. 1

The recent proliferation of political and social forums has given rise to a wealth of freely accessible naturalistic arguments. People can “talk ” to anyone they want, at any time, in any location, about any topic. Here we use a Mechanical Turk annotated corpus of forum discussions as a gold standard for the recognition of disagreement in online ideological forums. We analyze the utility of meta-post features, contextual features, dependency features and word-based features for signaling the disagreement relation. We show that using contextual and dialogic features we can achieve accuracies up to 68 % as compared to a unigram baseline of 63%. 1

Recent work on evaluativity or sentiment in the language sciences has focused on the contributions that lexical items provide. In this paper, we discuss contextual evaluativity, stance that is inferred from lexical meaning and pragmatic environments. Focusing on assessor-grounding claims like We liked him because he so clearly disliked Margaret Thatcher, we build a corpus and construct a system employing compositional principles of evaluativity calculation to derive that we dislikes Margaret Thatcher. The resulting system has an F-score of 0.90 on our dataset, outperforming reasonable baselines, and indicating the viability of inferencing in the evaluative domain. 1 Contextual Evaluativity A central aim of contemporary research on sentiment or evaluative language is the extraction of evaluative triples: 〈evaluator, target, evaluation〉. To date, both formal (e.g., Martin and White 2005, Potts 2005) and computational approaches (e.g., Pang and Lee 2008) have focused on how such triples are lexically encoded (e.g., the negative affect of scoundrel or dislike). While lexical properties are a key source of evaluative information, word-based considerations alone can miss pragmatic inferences resulting from context. (1), for example, communicates that the referent of we bears not only positive stance

Abstract not found