Unsupervised sentiment analysis with emotional signals

By Xia Hu, Jiliang Tang, Huiji Gao, Huan Liu. The Web Conference, 2013

418 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/5b34a69498c8a62e6f58b8150ec4452f9e8316c6

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

Predicting Polarities of Tweets by Composing Word Embeddings with Long Short-Term Memory (2015, background): “While some other works utilize the microblogging features uncommon in the formal literature, such as hashtags, emoticons (Hu et al., 2013a; Liu et al., 2012).”
Sentiment Embeddings with Applications to Sentiment Analysis (2016, methodology): “We use the carefully selected positive and negative emoticons [54](4), and save the tweets only containing positive emoticons or negative emoticons(5).”
Approaches, Tools and Applications for Sentiment Analysis Implementation (2015, background): “For example slang used in Social Networks is rarely supported in lexical methods [33].”
Opinion Mining and Information Fusion: A survey (2016, methodology): “The second corresponds to the use of Web crawlers in order to scrape the data from the desired websites [26, 27, 28].”
Learning Sentiment-Specific Word Embedding for Twitter Sentiment Classification (2014, methodology): “Each convolutional layer 1We use the emoticons selected by Hu et al. (2013).”
Ratings meet reviews, a combined approach to recommend (2014, background): “In another line of research called sentiment analysis [6, 12], a positive or negative label rather than an integer score is learnt for a short text.”
Evaluation Datasets for Twitter Sentiment Analysis: A survey and a new dataset, the STS-Gold (2013, methodology): “The OMD dataset is a popular dataset, which has been used to evaluate various supervised learning methods [10,23,21], as well as unsupervised methods [9] for polarity classification of tweets.”
Comparing and combining sentiment analysis methods (2013, neutral): “In ACM Web Science Conference (WebSci), 2011.”
Like It or Not (2016, methodology): “This dataset has been used by many researchers for TSA [Hu et al. 2013a, 2013b; Saif et al. 2012b; Speriosu et al. 2011].”
Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm (2017, methodology): “To analyze the effect of using a diverse emoji set we create a subset of our pretraining data containing tweets with one of 8 emojis that are similar to the positive/negative emoticons used by Tang et al. (2014) and Hu et al. (2013).”
On Stopwords, Filtering and Data Sparsity for Sentiment Analysis of Twitter (2014, background): “Stopwords, by definition, are meaningless words that have low discrimination power (Lo et al., 2005). The earliest work on stopwords removal is attributed to Hans Peter Luhan (1957), who suggested that words in natural language texts can be divided into keyword terms and non-keyword terms.”
Do users rate or review?: boost phrase-level sentiment labeling with review-level sentiment classification (2014, result): “The proof of the updating rule as well as its convergence is similar to the KKT condition approach in [2].”