NRC Affect Intensity Lexicon




The NRC Affect Intensity Lexicon provides real-valued affect intensity scores for four basic emotions (anger, fear, sadness, joy). We will be adding entries for four more emotions, as well as, valence, arousal, and dominance, shortly. (Access various other word-emotion, word-sentiment, and word-colour lexicons here.)

Email: saif.mohammad@nrc-cnrc.gc.ca



Words can be associated with different intensities (or degrees) of an emotion. For example, most people will agree that the word condemn is associated with a greater degree of anger (or more anger) than the word irritate. However, annotating instances for fine-grained degrees of affect is a substantially more difficult undertaking than categorical annotation: respondents are presented with greater cognitive load and it is particularly hard to ensure consistency (both across responses by different annotators and within the responses produced by the same annotator). We created an affect intensity lexicon with real-valued scores of association using best--worst scaling. We refer to this lexicon as the NRC Affect Intensity Lexicon. You can access a copy for non-commercial use here. (See terms of use at the bottom of this page.)

For a given word w and emotion e, the scores range from 0 to 1.

Details of the lexicon are in this paper:

Word Affect Intensities. Saif M. Mohammad. arXiv preprint arXiv:1704.08798, April 2017.
Paper (pdf)   

The lexicon has close to 6,000 entries for four basic emotions: anger, fear, joy, and sadness. (We will soon be adding entries for four more emotions: trust, disgust, anticipation, and surprise. We will also be adding entries for valence, arousal, and dominance.) It includes common English terms as well as terms that are more prominent in social media platforms, such as Twitter. It includes terms that are associated with emotions to various degrees. For a given emotion, this even includes some terms that may not predominantly convey that emotion (or that convey an antonymous emotion), and yet tend to co-occur with terms that do. (Antonymous terms tend to co-occur with each other more often than chance, and are particularly problematic when one uses automatic co-occurrence-based statistical methods to capture word--emotion connotations.) Example entries from the lexicon are shown below.

 

Terms of use:
  • The lexicons mentioned in this page are available for direct download and can be used freely for research purposes.
  • The papers listed next to the lexicons provide details of the creation and use. If you use a lexicon, then please cite the associated papers.
  • If interested in commercial use of any of these lexicons, send email to the contact.
  • If you use a lexicon in a product or application, then please credit the authors and NRC appropriately. Also, if you send us an email, we will be thrilled to know about how you have used the lexicon.
  • Rather than redistributing the data, please direct interested parties to this page.
  • National Research Council Canada (NRC) disclaims any responsibility for the use of the lexicons listed here and does not provide technical support. However, the contact listed above will be happy to respond to queries and clarifications.
 
We will be happy to hear from you, especially if:
  • you give us feedback regarding these lexicons.
  • you tell us how you have (or plan to) use the lexicons.
  • you are interested in having us analyze your data for sentiment, emotion, and other affectual information.
  • you are interested in a collaborative research project. We also regularly hire graduate students for research internships.