Summary of the paper

Title GermanPolarityClues: A Lexical Resource for German Sentiment Analysis
Authors Ulli Waltinger
Abstract In this paper, we propose GermanPolarityClues, a new publicly available lexicalresource for sentiment analysis for the German language. While sentimentanalysis and polarity classification has been extensively studied at differentdocument levels (e.g. sentences and phrases), only a few approaches exploredthe effect of a polarity-based feature selection and subjectivity resources forthe German language. This paper evaluates four different English and threedifferent German sentiment resources in a comparative manner by combining apolarity-based feature selection with SVM-based machine learning classifier. Using a semi-automatic translation approach, we were able to construct threedifferent resources for a German sentiment analysis. The manually finalizedGermanPolarityClues dictionary offers thereby a number of 10, 141 polarityfeatures, associated to three numerical polarity scores, determining thepositive, negative and neutral direction of specific term features. While theresults show that the size of dictionaries clearly correlate to polarity-basedfeature coverage, this property does not correlate to classification accuracy.Using a polarity-based feature selection, considering a minimum amount of priorpolarity features, in combination with SVM-based machine learning methodsexhibits for both languages the best performance (F1: 0.83-0.88).
Language Knowledge Discovery/Representation
Topics Emotion Recognition/Generation, Information Extraction, Information Retrieval, Knowledge Discovery/Representation
Full paper GermanPolarityClues: A Lexical Resource for German Sentiment Analysis
Bibtex @InProceedings{WALTINGER10.91,
  author = {Ulli Waltinger},
  title = {GermanPolarityClues: A Lexical Resource for German Sentiment Analysis},
  booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)},
  year = {2010},
  month = {may},
  date = {19-21},
  address = {Valletta, Malta},
  editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias},
  publisher = {European Language Resources Association (ELRA)},
  isbn = {2-9517408-6-7},
  language = {english}
 }
Powered by ELDA © 2010 ELDA/ELRA