Using Emoticons to Reduce Dependency in Machine Learning Techniques for Sentiment Classification

By Jonathon Read. Annual Meeting of the Association for Computational Linguistics, 2005

643 citations recorded by the source corpus.

Sentiment Classification seeks to identify a piece of text according to its author's general feeling toward their subject, be it positive or negative. Traditional machine learning techniques have been applied to this problem with reasonable success, but they have been shown to work well only when there is a good match between the training and test data with respect to topic. This paper demonstrates that match with respect to domain and time is also important, and presents preliminary experiments with training data labeled with emoticons, which has the potential of being independent of domain, topic and time.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/1b7c02e4f07477a5ae2d1aa52db8ac6ddb55f53c

How researchers see this paper

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

In Artificial Intelligence

In Information Systems

Citations in the corpus

Twitter as a Corpus for Sentiment Analysis and Opinion Mining (2010, methodology): “To collect negative and positive sentiments, we followed the same procedure as in (Read, 2005; Go et al., 2009).”
EmoNet: Fine-Grained Emotion Detection with Gated Recurrent Neural Networks (2017, methodology): “, 2007) and for other semantic tasks such as sentiment analysis (Read, 2005; Go et al., 2009) and sarcasm detection (González-Ibánez et al.”
Ontology-based sentiment analysis of twitter posts (2013, methodology): “Working with noisy labels or distant supervision (Read, 2005), by creating, for example, emoticon vocabularies for representing sentiment and for training supervised sentiment classifiers, such as Naïve Bayes (NB), Maximum Entropy (MaxEnt) and Support Vector Machines (SVMs) (Barbosa & Feng, 2010; Pa”
Twitter Sentiment Classiﬁcation using Distant Supervision (2009, positive): “Read [10] shows that using emoticons as labels for positive and sentiment is effective for reducing dependencies in machine learning techniques.”
Sentiment of Emojis (2015, methodology): “In an early work [10], a basic distinction between positive and negative emoticons was used to automatically generate positive and negative samples of texts.”
CARER: Contextualized Affect Representations for Emotion Recognition (2018, methodology): “…we obtain Twitter-based, pre-trained word embeddings from (Deriu et al., 2017) and reweight them via a sentiment corpus through distant supervision (Read, 2005; Go et al., 2009).7 We trained a fully connected neural network (1 hidden layer) with 10 epochs via backpropagation as in (Deriu et al.,…”
Like It or Not (2016, methodology): “Go et al. used the technique that was demonstrated by Read [2005] to collect the data.”
TOM: Twitter opinion mining framework using hybrid classification scheme (2014, methodology): “The research paper introduced a system, MOATweetReader, which processes the tweets in real-time despite their dynamic nature.”
User-level sentiment analysis incorporating social networks (2011, methodology): “Variation in SVM training We now briefly mention our experiments with two alternative training sets for the tweet-level SVM that underlies the SVM vote baseline: (a) a single set of outof-domain tweets labeled using emoticons as distant supervision [17]; (b) the same 5 topical sets described in §5.”
Co-Training for Cross-Lingual Sentiment Classification (2009, background): “Since the work of Pang et al. (2002), various classification models and linguistic features have been proposed to improve the classification performance (Pang and Lee, 2004; Mullen and Collier, 2004; Wilson et al., 2005; Read, 2005).”
Posterior calibration and exploratory analysis for natural language processing models (2015, methodology): “We collect a dataset consisting of tweets identiﬁed by the Twitter API as English, collected from 2014 to 2015, with the “emoticon trick” (Read, 2005; Lin and Kolcz, 2012) to label tweets that contain at least one occurrence of the smiley emoticon “:)” as “happy” ( y = 1 ) and others as y = 0 .”
Sentiment in short strength detection informal text (2010, neutral): “Nevertheless, with a few exceptions (Read, 2005; Wilson et al., 2006), explicit comparisons with other methods have not been included in opinion mining publications.”
Approaches, Tools and Applications for Sentiment Analysis Implementation (2015, methodology): “Therefore, emoticons have been often used in combination with other techniques for building a training dataset in supervised machine learning techniques [36].”