KLUE: simple and robust methods for polarity classification

From Brede Wiki
Jump to: navigation, search
Conference paper (help)
KLUE: simple and robust methods for polarity classification
Authors: Thomas Proisl, Paul Greiner, Stefan Evert, Besim Kabashi
Citation: Seventh International Workshop on Semantic Evaluation (SemEval 2013) 2  : 395-401. 2013
Editors:
Publisher: Association for Computational Linguistics
Meeting: Second Joint Conference on Lexical and Computational Semantics
Database(s):
DOI: Define doi.
Link(s): http://www.aclweb.org/anthology-new/S/S13/S13-2065.pdf
Search
Web: DuckDuckGo Bing Google Yahoo!Google PDF
Article: Google Scholar PubMed
Restricted: DTU Digital Library
Services
Format: BibTeX

KLUE: simple and robust methods for polarity classification describes a method for a SemEval-2013 task on sentiment analysis of Twitter posts and SMS text messages.

The AFINN word list is used and extended from distributional semantic models. This model was constructed from Wikipedia via Wackypedia:

http://wacky.sslmit.unibo.it/

A reimplementation of the method is described in Webis: an ensemble for Twitter sentiment detection.

[edit] Results

Task B Twitter test set accuracy: (best): 0.7031

Task B SMS test set accuracy: (best): 0.7259

Task A Twitter test set accuracy: (best): 0.7493

Task A SMS test set accuracy: (best): 0.6829

The results are compared to the results of other research groups on the same data set in the paper: SemEval-2013 task 2: sentiment analysis in Twitter.

KLUE performed overall reasonable well. In one task the KLUE system even was number 3 out of 30.

[edit] Related papers

  1. BOUNCE: sentiment classification in Twitter using rich feature sets
  2. SemEval-2013 task 2: sentiment analysis in Twitter
  3. senti.ue-en: an approach for informally written short texts in SemEval-2013 sentiment analysis task
  4. Webis: an ensemble for Twitter sentiment detection
Personal tools