The Community for Technology Leaders
RSS Icon
Issue No.04 - July/August (2008 vol.23)
pp: 43-50
V.S. Subrahmanian , University of Maryland
Diego Reforgiato , University of Maryland
Most research on determining the strength of subjective expressions in a sentence or document uses single, specific parts of speech such as adjectives, adverbs, or verbs. To date, almost no research covers the development of a single comprehensive framework in which we can analyze sentiment that takes all three into account. The authors propose the AVA (adjective verb adverb) framework for identifying opinions on any given topic. In AVA, a user can select any topic t of interest and any document d. AVA will return a score that d expresses topic t. The score is expressed on a –1 (maximally negative) to +1 (maximally positive) scale.
opinion analysis, sentiment analysis, computational cultural dynamics
V.S. Subrahmanian, Diego Reforgiato, "AVA: Adjective-Verb-Adverb Combinations for Sentiment Analysis", IEEE Intelligent Systems, vol.23, no. 4, pp. 43-50, July/August 2008, doi:10.1109/MIS.2008.57
1. B. Levin, English Verb Classes and Alternations: A Preliminary Investigation, Univ. of Chicago Press, 1993.
2. F. Benamara et al., "Sentiment Analysis: Adverbs and Adjectives Are Better than Adverbs Alone," Proc. 2007 Int'l Conf. Weblogs and Social Media (ICWSM07), 2007; .
3. C. Cesarano et al., "OASYS: An Opinion Analysis System," Proc. AAAI 2006 Spring Symp. Computational Approaches to Analyzing Weblogs, AAAI Press, 2006, pp. 21–26.
4. P. Turney, "Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews," Proc. 40th Ann. Meeting Assoc. for Computational Linguistics (ACL 02), Assoc. for Computational Linguistics, 2002, pp. 417–424.
5. S.O. Kim and E. Hovy, "Determining the Sentiment of Opinions," Proc. 20th Int'l Conf. Computational Linguistics (COLING04), ACM Press, 2004, pp. 1367–1373.
30 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool