Sentiment strength detection in short informal text

From Brede Wiki
Jump to: navigation, search
Paper (help)
Sentiment strength detection in short informal text
Authors: Mike Thelwall, Kevan Buckley, Georgios Paltoglou, Di Cai, Arvid Kappas
Citation: Journal of the American Society for Information Science and Technology 61 (12): 2544-2558. 2010 December
Database(s): Google Scholar cites
DOI: 10.1002/asi.21416.
Link(s):
Search
Web: Bing Google Yahoo!Google PDF
Article: BASE Google Scholar PubMed
Restricted: DTU Digital Library
Other: NIF
Services
Format: BibTeX
Extract:

Sentiment in short strength detection informal text (the title should have been Sentiment strength detection in short informal text: there is an errata to the text.) reports the development of a sentiment analysis system for estimating the sentiments in short texts (MySpace comments). The authors call the system SentiStrength.

The researchers set 5 human coders to label MySpace comments for sentiment. Positive and negative sentiment was labeled independently on two 5 point scales.

Features:

  1. Word list
  2. Score adjustment
  3. "miss" word
  4. Spelling correction
  5. Booster words
  6. Negation
  7. Spelling boosting
  8. Emoticon
  9. Exclamation
  10. Questions

Furthermore these features were considered:

  1. Phrase identification
  2. Semantic disambiguation

The SentiStrength algorithm was compared with "a range of standard machine-learning classification algorithms in Weka (Witten & Frank, 2005) using the frequencies of each word in the sentiment word list as the feature set." (page 2550).


[edit] Results

  • They found a Pearson correlation coeffients on 0.639-0.664 for the agreement between 3 human coders of sentiment strength on 1,041 MySpace comments.

[edit] Related studies

  1. A new ANEW: evaluation of a word list for sentiment analysis in microblogs
  2. Micro-blogging sentiment detection by collaborative online learning
  3. Recognizing contextual polarity: an exploration of features for phrase-level sentiment analysis also considers short text sentiment analysis
  4. Robust sentiment detection on Twitter from biased and noisy data
  5. Sentiment in Twitter events is a newer study by the first author, where SentiStrength is used for Twitter sentiment analysis.
Personal tools