Message Boards Message Boards

Perform tone analysis of texts?

Hi, We use IBM Watson for Tone Analysis of text, we wanted to enhance the same in mathamatica, as we would be extending the scope of emotional analysis. Has anybody in the community built something similar to IBM Tone Analysis, if so would love to see the approach and collaborate.

Warm Regards Jayanth
POSTED BY: Jayanth Bagare
9 Replies
Sort By:

Hi Jaynath, out of curiosity, what domain of texts are you interested in analysing?
POSTED BY: Arno Bosse

Hello Arno We are looking at analysis of content specifically documents blogs etc. Trying to classify it based on Robert plutchik model of emotional analysis
POSTED BY: Jayanth Bagare

In addition it is in the area of marketing campaigns as well as employee engagement
POSTED BY: Jayanth Bagare

Perhaps this could also be useful reading for you: Market Sentiment Analysis
POSTED BY: Vitaliy Kaurov

Thanks really helps
POSTED BY: Jayanth Bagare

of course you can simply use Classify with your examples and outcomes to train a classifier! Look at the documentation of Classify it has (simple) examples that do such thing.
POSTED BY: Sander Huisman

Thanks I guess that is the approach I will follow now as I have a labelled data set
POSTED BY: Jayanth Bagare

There is such a function built-in:

Classify["Sentiment", {"I love this movie", "I am so sad", 
  "My phone broke again", "He's amazing", "That is terrible news", 
  "That is terrible news!"}]

But, as you can, it has its shortcomings, as per the last example...
POSTED BY: Sander Huisman

thanks for sharing the classify example. I was wondering if we could use Wordnet databases to classify the same
POSTED BY: Jayanth Bagare
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Tag limit exceeded
Note: Only the first five people you tag will receive an email notification; the other tagged names will appear as links to their profiles.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract