Sentiment strength detection for the social web

By Mike Thelwall, Kevan Buckley, Georgios Paltoglou. J. Assoc. Inf. Sci. Technol., 2012

1,154 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/d02927d4de4a2a51cced4970da04b812cbee4342

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

Citations in the corpus

Refining Word Embeddings for Sentiment Analysis (2017, background): “In addition to the E-ANEW, other lexicons such as Sen-tiWordNet (Esuli and Fabrizio, 2006), SoCal (Taboada et al., 2011), SentiStrength (Thelwall et al., 2012), Vader (Hutto et al., 2014), ANTUSD (Wang and Ku, 2016) and SCL-NMA (Kiritchenko and Mohammad, 2016) also provide real-valued sentiment…”
Bug report, feature request, or simply praise? On automatically classifying app reviews (2015, background): “SentiStrength [32] assigns for each review one negative sentiment score in a scale of -5 to -1 and one positive score in a scale of 1 to 5.”
New classification models for detecting Hate and Violence web content (2015, background): “SentiStrength (Thelwall et al, 2010, 2012) takes a lexical approach to sentiment analysis, making use of a combination of sentiment lexical resources, semantic rules, heuristic rules and additional rules.”
A Survey on Text Classification: From Traditional to Deep Learning (2020, methodology): “The SA datasets include Movie Review (MR) [257] [226], Stanford Sentiment Treebank (SST) [227], Multi-Perspective Question Answering (MPQA) [258] [229], IMDB [230], Yelp [231], Amazon Reviews (AM) [93], NLP&CC 2013 [111], Subj [250], CR [251], SS-Twitter [259], SS-Youtube [259], SE1604 [260] and so ”
A Survey on Text Classification: From Shallow to Deep Learning (2020, neutral): “F 5 93 3,650,000 [5] [46] [131] [190] SA DSTC 4 89 - 30,000 [9] [191] DAC MRDA 5 - 62,000 [9] [192] DAC SwDA 43 - 1,022,000 [9] [193] DAC RCV1 103 240 807,595 [117] [123] [68] [134] [139] [194] NC RCV1-V2 103 124 804,414 [74] [139] [195] NC NLP&CC 2013 2 - 115,606 [125] [126] SA SS-Twitter 2 - 2,113”
The Validity of Sentiment Analysis: Comparing Manual Annotation, Crowd-Coding, Dictionary Approaches, and Machine Learning Algorithms (2021, background): “Similarly, machine learning models that are optimized for a certain task (such as distinguishing positive film reviews from negative ones) can give misleading results for social science research by identifying spurious patterns in the data that was used to train these algorithms (Thelwall et al., 20”
How Do Users Like This Feature? A Fine Grained Sentiment Analysis of App Reviews (2014, background): “Previous research has shown that SentiStrength has a good accuracy for short texts in social media, such as, Twitter and movie reviews [27].”
Explainable recommendation with fusion of aspect information (2018, background): “They explore user opinion [32, 35, 39, 40] from reviews by means of a sentiment lexicon [8, 36, 37] or corpus [14, 28, 44] and integrate these opinion information into recommendations.”
Learning Sentiment-Specific Word Embedding for Twitter Sentiment Classification (2014, methodology): “The lexicon-based approaches (Turney, 2002; Ding et al., 2008; Taboada et al., 2011; Thelwall et al., 2012) mostly use a dictionary of sentiment words with their associated sentiment polarity, and incorporate negation and intensification to compute the sentiment polarity for each sentence (or…”
The evolution of sentiment analysis - A review of research topics, venues, and top cited papers (2016, methodology): “In 2012, Thelwall, Buckley and Paltoglou [Thelwall et al. 2012] evaluate a tool named “SentiStrength 2”, which is a lexicon based classifier used together with various machine learning algorithms.”
Positive, Negative, or Neutral: Learning an Expanded Opinion Lexicon from Emoticon-Annotated Tweets (2015, methodology): “The first collection is 6HumanCoded9, which was used to evaluate SentiStrength [Thelwall et al., 2012].”
Contextual semantics for sentiment analysis of Twitter (2016, methodology): “25 38.95 SentiStrength 62.07 64.15 91.89 75.56 40.00 9.52 15.38 52.08 50.71 51.38 Senti-Median (SentiWordNet) 81.03 90.91 78.95 84.51 68.00 85.00 75.56 79.45 81.97 80.03 Senti-Median (MPQA) 77.59 90.00 72.97 80.60 64.29 85.71 73.47 77.14 79.34 77.03 Senti-Median (Thelwall-Lexicon) 79.31 84.85 80.00 ”
Like It or Not (2016, methodology): “The algorithm was extended by Thelwall et al. [2012] by introducing idiom lists and new sentiment words in the lexicon and by strength boosting using emphatic lengthening.”
Social Collaborative Viewpoint Regression with Explainable Recommendations (2017, methodology): “Lexicon-based methods [11, 43, 45] utilize a lexicon of sentiment words to predict sentiment labels, whereas corpus-based methods [16, 29, 55] classify sentences to sentiment polarities using corpora that are labeled with sentiment labels.”
Evaluation Datasets for Twitter Sentiment Analysis: A survey and a new dataset, the STS-Gold (2013, methodology): “Instead of the final sentiment labels associated to the tweets, some datasets provide a numeric sentiment strength between -5 and 5 defining a range from negative to positive polarity [24,25].”