Sparse, Contextually Informed Models for Irony Detection: Exploiting User Communities, Entities and Sentiment
Published 1 January 2015Open access
Byron Wallace, Do Kook Choe, Eugene Charniak
Citations91
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
It is shown that this approach improves verbal irony classification performance and is proposed a mixed regularization strategy that places a sparsity-inducing `1 penalty on the contextual feature weights on top of the `2 penalty applied to all model coefficients.
Abstract
Byron C. Wallace, Do Kook Choe, Eugene Charniak. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.
Keywords
Computer Science
arXiv (Cornell University)Scikit-learn: Machine Learning in Python
63,264 Citations2012Fabián Pedregosa, Gaël Varoquaux +17 more
Journal of the Royal Statistical Society Series B (Statistical Methodology)Regularization and Variable Selection Via the Elastic Net
20,982 Citations2005Hui Zou, Trevor Hastie
Lecture notes in computer scienceText categorization with Support Vector Machines: Learning with many relevant features
7,925 Citations1998Thorsten Joachims
SVMs achieve substantial improvements over the currently best performing methods and behave robustly over a variety of di-erent learning tasks, eliminating the need for manual parameter tuning.
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
6,774 Citations2013Richard Socher, Alex Perelygin +5 more
A Sentiment Treebank that includes fine grained sentiment labels for 215,154 phrases in the parse trees of 11,855 sentences and presents new challenges for sentiment compositionality, and introduces the Recursive Neural Tensor Network.
Feature-rich part-of-speech tagging with a cyclic dependency network
2,851 Citations2003Kristina Toutanova, Dan Klein +2 more
A new part-of-speech tagger is presented that demonstrates the following ideas: explicit use of both preceding and following tag contexts via a dependency network representation, broad use of lexical features, and effective use of priors in conditional loglinear models.
Identifying Sarcasm in Twitter: A Closer Look
632 Citations2011Roberto González‐Ibáñez, Smaranda Muresan +1 more
This work reports on a method for constructing a corpus of sarcastic Twitter messages in which determination of the sarcasm of each message has been made by its author and uses this reliable corpus to compare sarcastic utterances in Twitter to utterances that express positive or negative attitudes without sarcasm.
Sarcasm as Contrast between a Positive Sentiment and Negative Situation
591 Citations2013Ellen Riloff, Ashequl Qadir +4 more
This work develops a sarcasm recognizer that automatically learns lists of positive sentiment phrases and negative situation phrases from sarcastic tweets and shows that identifying contrasting contexts using the phrases learned through bootstrapping yields improved recall for sarcasm recognition.
Language Resources and EvaluationA multidimensional approach for detecting irony in Twitter
459 Citations2012Antonio Reyes, Paolo Rosso +1 more
A new model of irony detection that is assessed along two dimensions: representativeness and relevance is constructed, and initial results are largely positive, and provide valuable insights into the figurative issues facing tasks such as sentiment analysis, assessment of online reputations, or decision making.
Semi-supervised recognition of sarcastic sentences in Twitter and Amazon
422 Citations2010Dmitry Davidov, Oren Tsur +1 more
This paper experiments with semi-supervised sarcasm identification on two very different data sets: a collection of 5.9 million tweets collected from Twitter, and aCollection of 66000 product reviews from Amazon.
Proceedings of the International AAAI Conference on Web and Social MediaICWSM — A Great Catchy Name: Semi-Supervised Recognition of Sarcastic Sentences in Online Product Reviews
352 Citations2010Oren Tsur, Dmitry Davidov +1 more
A novel Semi-supervised Algorithm for Sarcasm Identification that recognizes sarcastic sentences in product reviews and speculate on the motivation for using sarcasm in online communities and social networks is presented.
FigshareSparse Additive Generative Models of Text
289 Citations2018Jacob Eisenstein, Amr Ahmed +1 more
This approach has two key advantages: it can enforce sparsity to prevent overfitting, and it can combine generative facets through simple addition in log space, avoiding the need for latent switching variables.
Clues for detecting irony in user-generated contents
240 Citations2009Paula Carvalho, Luís Sarmento +2 more
It is shown that it is possible to find ironic sentences with relatively high precision by exploring certain oral or gestural clues in user comments, such as emoticons, onomatopoeic expressions for laughter, heavy punctuation marks, quotation marks and positive interjections.
Stochastic gradient descent training for L1-regularized log-linear models with cumulative penalty
229 Citations2009Yoshimasa Tsuruoka, Jun’ichi Tsujii +1 more
Experimental results demonstrate that the proposed SGD method can produce compact and accurate models much more quickly than a state-of-the-art quasi-Newton method for L1-regularized loglinear models.
Irony and Sarcasm: Corpus Generation and Analysis Using Crowdsourcing
180 Citations2012Elena Filatova
A corpus generation experiment is described where regular and sarcastic Amazon product reviews are collected and the resulting corpus can be used for identifying sarcasm on two levels: a document and a text utterance.
yeah right: sarcasm recognition for spoken dialogue systems
154 Citations2006Joseph Tepperman, David Traum +1 more
The results demonstrate that spectral and contextual features can be used to detect sarcasm as well as a human annotator would, and confirm a long-held claim in the field of psychology – that prosody alone is not sufficient to discern whether a speaker is being sarcastic.
Automatic satire detection
147 Citations2009Clint Burfoot, Timothy Baldwin
The novel task of determining whether a newswire article is "true" or satirical is introduced, and SVMs, feature scaling, and a number of lexical and semantic feature types are experimented on.
Humans Require Context to Infer Ironic Intent (so Computers Probably do, too)
130 Citations2014Byron Wallace, Do Kook Choe +2 more
It is shown that annotators frequently require context to make judgements concerning ironic intent, and that machine learning approaches tend to misclassify those same comments for which annotators required additional context.
FigshareDiscovering Sociolinguistic Associations with Structured Sparsity
124 Citations2018Jacob Eisenstein, Noah A. Smith +1 more
A method to discover robust and interpretable sociolinguistic associations from raw geotagged text data is presented, using aggregate demographic statistics about the authors' geographic communities to solve a multi-output regression problem between demographics and lexical frequencies.
Artificial Intelligence ReviewComputational irony: A survey and new perspectives
87 Citations2013Byron Wallace
It is argued that current machine learning methods rely too heavily on shallow, unstructured, syntactic modeling of text to consistently discern ironic intent, and that without an internal model of the speaker, irony detection is hopeless.
Linguistic Structured Sparsity in Text Categorization
63 Citations2014Dani Yogatama, Noah A. Smith
Three linguistically motivated structured regularizers based on parse trees, topics, and hierarchical word clusters for text categorization are introduced, showing that these regularizers consistently improve classification accuracies compared to standard regularizers that penalize features in isolation.
Really? Well. Apparently Bootstrapping Improves the Performance of Sarcasm and Nastiness Classifiers for Online Dialogue
45 Citations2013Stephanie M. Lukin, Marilyn Walker
A bootstrapping method to train classifiers to identify two different types of subjective language in dialogue: sarcasm and nastiness, originally proposed in a monologic domain is tested.
FigshareMaking the Most of Bag of Words: Sentence Regularization with Alternating Direction Method of Multipliers
39 Citations2018Dani Yogatama, Noah A. Smith
This work applies the sparse overlapping group lasso with one group for every bundle of features occurring together in a training-data sentence, leading to thousands to millions of overlapping groups, to efficiently solve the resulting optimization challenge using the alternating directions method of multipliers.