Random observations on Sentiment Analysis

  • with S.A., context is everything
  • S.A. does well with web pages
  • S.A. does nothing for Tweets (beyond the very basic smiley detection)
  • computers don’t understand sarcasm
  • “positive”, “negative”, and “neutral” are not enough for business purposes
  • “negative” sentiment may indicate a positive opportunity, and vice versa (again, industry context)


Btw, after recently analyzing 1,000,000 web objects related to a particular topic, we noticed that regardless of the vendor, “Legal Disclaimer”, “Terms & Conditions”, and “Warranty” documents would almost universally be tagged as Negative.  Makes sense.

Technologist, parallel entrepreneur. Interests: travel, photography, big data, analytics, predictive modeling.

Tagged with: , , , , , , , ,
Posted in Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: