Unveiling factors influencing judgment variation in sentiment analysis with natural language processing and statistics

HIGHLIGHTS

• What: This study explores three important factors influencing variation in crowdsourced polarity judgments focusing on TripAdvisor reviews in Spanish. Main dataset=one word titles (ShortT); additional datasets for comparison=titles without negation (PosT); titles with negation (NegT) Factors influencing crowdsourced polarity judgements with NLP and statistics Methods for testing Hypotheses Part of the methodology the authors use in this paper is based on the preliminary study in which the authors have shown that one-word titles have the smallest ambiguity in comparison to other datasets . The authors focus on POS tags corresponding to content words in Stanza . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.