Are Word Embedding-based Features Useful for Sarcasm Detection?

@inproceedings{Joshi2016AreWE,
  title={Are Word Embedding-based Features Useful for Sarcasm Detection?},
  author={Aditya Joshi and Vaibhav Tripathi and Kevin Patel and Pushpak Bhattacharyya and Mark James Carman},
  booktitle={EMNLP},
  year={2016}
}
This paper makes a simple increment to state-of-the-art in sarcasm detection research. Existing approaches are unable to capture subtle forms of context incongruity which lies at the heart of sarcasm. We explore if prior work can be enhanced using semantic similarity/discordance between word embeddings. We augment word embedding-based features to four feature sets reported in the past. We also experiment with four types of word embeddings. We observe an improvement in sarcasm detection… 

Tables from this paper

The Significance of Global Vectors Representation in Sarcasm Analysis
TLDR
The word vector model (GloVe) that integrate two major model families; Global statistics of matrix factorization technique and local context window based approach is investigated, able to build a word representation that do not only learn semantics and grammar information, but also captures the context of the word and information of global corpus statistics.
Topic-Enriched Word Embeddings for Sarcasm Identification
TLDR
The experimental analysis indicate that topic-enriched word embedding schemes utilized in conjunction with conventional feature sets can yield promising results for sarcasm identification.
IA-BERT: context-aware sarcasm detection by incorporating incongruity attention layer for feature extraction
TLDR
IA-BERT is a model architecture that considers contextual information to identify incongruity features that lie in sarcastic texts and leverages BERT pretrained embedding and yields a performance improvement from the standard fine-tuned BERT classifier.
A Contextual Word Embedding for Arabic Sarcasm Detection with Random Forests
TLDR
This paper proposes a new approach for improving Arabic sarcasm detection using data augmentation, contextual word embedding and random forests model to get the best results.
Sarcasm Detection with Self-matching Networks and Low-rank Bilinear Pooling
TLDR
This work proposes a novel self-matching network to capture sentence ”incongruity” information by exploring word-to-word interactions and incorporates a bi-directional LSTM network into the proposed network to retain compositional information.
Automatic Sarcasm Detection
Automatic sarcasm detection is the task of predicting sarcasm in text. This is a crucial step to sentiment analysis, considering prevalence and challenges of sarcasm in sentiment-bearing text.
Affective Representations for Sarcasm Detection
TLDR
A novel model for automated sarcasm detection in text, called Affective Word Embeddings for Sarcasm (AWES), which incorporates affective information into word representations is introduced, which demonstrates the effectiveness of the proposed model.
Evaluation of Different Sarcasm Detection Models for Arabic News Headlines
TLDR
Bridging the gap in sarcasm detection of the Arabic language is the target behind this work, and efficient models in short text classification are tested for detecting sarcasm in the Arabic news headlines for the first time.
Learning to capture contrast in sarcasm with contextual dual-view attention network
TLDR
A contextual dual-view attention network (CDVaN) is proposed for sarcasm detection according to the formation mechanism of sarcasm and a Contrast Understanding Unit is proposed to effectively extract the contrast between the positive sentiment and the negative situation from the view of formation mechanism.
A2Text-Net: A Novel Deep Neural Network for Sarcasm Detection
TLDR
A new deep neural network is employed: A2Text-Net to mimic the face-to-face speech, which integrates auxiliary variables such as punctuations, part of speech (POS), numeral, emoji, etc. to increase classification performance.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 30 REFERENCES
Automatic Sarcasm Detection
Automatic sarcasm detection is the task of predicting sarcasm in text. This is a crucial step to sentiment analysis, considering prevalence and challenges of sarcasm in sentiment-bearing text.
ICWSM - A Great Catchy Name: Semi-Supervised Recognition of Sarcastic Sentences in Online Product Reviews
TLDR
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.
Sarcastic or Not: Word Embeddings to Predict the Literal or Sarcastic Meaning of Words
TLDR
This work re-frame the sarcasm detection task as a type of word sense disambiguation problem, where the sense of a word is either literal or sarcastic, and addresses two issues: how to collect a set of target words that can have either literal and sarcastic meanings depending on context.
Semi-supervised recognition of sarcastic sentences in Twitter and Amazon
TLDR
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.
Sparse, Contextually Informed Models for Irony Detection: Exploiting User Communities, Entities and Sentiment
TLDR
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.
An Impact Analysis of Features in a Classification Approach to Irony Detection in Product Reviews
TLDR
This paper forms the problem as a supervised classification task and evaluates different classifiers, reaching an F1-measure of up to 74 % using logistic regression.
The perfect solution for detecting sarcasm in tweets #not
TLDR
It is hypothesize that explicit markers such as hashtags are the digital extralinguistic equivalent of nonverbal expressions that people employ in live interaction when conveying sarcasm, and shows that sarcasm is often signalled by hyperbole, using intensifiers and exclamations; in contrast, non-hyperbolic sarcastic messages often receive an explicit marker.
Your Sentiment Precedes You: Using an author’s historical tweets to predict sarcasm
TLDR
This work presents the first quantitative evidence to show that historical tweets by an author can provide additional context for sarcasm detection, and uses a contrast-based and historical tweet-based approach.
Dependency-Based Word Embeddings
TLDR
The skip-gram model with negative sampling introduced by Mikolov et al. is generalized to include arbitrary contexts, and experiments with dependency-based contexts are performed, showing that they produce markedly different embeddings.
"sure, I Did the Right Thing": a System for Sarcasm Detection in Speech
TLDR
This paper presents an effective way of modeling and applying prosodic contours to categorical representations of pitch and intensity contours obtained via k-means clustering, and is able to predict sarcasm with 81.57% accuracy.
...
1
2
3
...