Detecting speculations, contrasts and conditionals in consumer reviews

• Maria Skeppstedt ^[1] ; Teri Schamp-Bjerede ^[2] ; Magnus Sahlgren ^[3] ; Carita Paradis ^[2] ; Andreas Kerren ^[1]
  1. [1] Linnaeus University

     Linnaeus University

     Suecia

     
  2. [2] Lund University

     Lund University

     Suecia

     
  3. [3] Gavagai AB, Stockholm, Sweden

  Mostrar afiliaciones +
• Localización: 6th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis WASSA 2015: Workshop Proceedings : 17 September 2015 Lisboa, Portugal / Alexandra Balahur Dobrescu (ed. lit.), Erik van der Goot (ed. lit.), Piek Vossen (ed. lit.), Andrés Montoyo Guijarro (ed. lit.), 2015, ISBN 978-1-941643-32-7, págs. 162-168
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • A support vector classifier was compared to a lexicon-based approach for the task of detecting the stance categories speculation, contrast and conditional in English consumer reviews. Around 3,000 train- ing instances were required to achieve a stable performance of an F-score of 90 for speculation. This outperformed the lexicon-based approach, for which an F- score of just above 80 was achieved. The machine learning results for the other two categories showed a lower average (an approximate F-score of 60 for contrast and 70 for conditional), as well as a larger variance, and were only slightly better than lexicon matching. Therefore, while machine learning was successful for detecting speculation, a well-curated lexicon might be a more suitable approach for detecting contrast and conditional.