Skip to content

Instantly share code, notes, and snippets.

@greeness

greeness/gist:1410425

Created November 30, 2011 19:28
Show Gist options
  • Save greeness/1410425 to your computer and use it in GitHub Desktop.
Save greeness/1410425 to your computer and use it in GitHub Desktop.
Download ZIP
sentiment analysis notes
Raw
gistfile1.md

Challenges

  • query classification: to tell if a text is a review or opinion

  • which documents or portions of documents contain review-like or opinionated material.

  • identifying the overall sentiment expressed

  • the system needs to present the sentiment information it has garnered in some reasonable summary fashion.

  • aggregation of "votes" that may re registered on different scales

  • selective highlighting of some opinions

  • representation of points of disagreement and points of consensus

  • identification of communities of opinion holders

  • accounting for different levels of authority among opinion holders.

  • visualization of sentiment data

  • context-sensitive and domain dependent

  • "go read the book" most likely indicates positive sentiment for book reviews, but negative sentient for movie reviews.

  • order-dependence

Problem Formulation and Key Concepts

Sentiment Polarity and Degrees of Positivity

Polarity Classification: The binary classification task of labeling an opinionated document as expressing either an over positive or an overall negative opinion.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment