Skip to content

Update README.rst#75

Open
ydineshy225 wants to merge 1 commit intocjhutto:masterfrom
ydineshy225:patch-1
Open

Update README.rst#75
ydineshy225 wants to merge 1 commit intocjhutto:masterfrom
ydineshy225:patch-1

Conversation

@ydineshy225
Copy link
Copy Markdown

I guess limits of compound values to categorize positive, negative and neutral should be like compound score >=0.5, (compound score < 0.5 and compound score > -0.5) and compound score <=-0.5. I don't find proper result with above mentioned limits could you please check on that.

I guess limits of compound values to categorize positive, negative and neutral should be like ``compound`` score >=0.5, (``compound`` score < 0.5 and ``compound`` score > -0.5) and ``compound`` score <=-0.5. I don't find proper result with above mentioned limits could you please check on that.
Copy link
Copy Markdown
Owner

@cjhutto cjhutto left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since sentiment scores are very closely tied to the actual text (and the domain of analysis for the data set), threshold values will be up to the modeler/analyst... use what works for you, but the thresholds recommended here (i.e., +/- 0.05) have a very "tight" allowance for what is considered neutral (and everything else is either positive to some [varying] degree of intensity, or negative to some degree).

We'll leave the recommended thresholds in the README as is, but I'll leave this note open so that these comments are visible to others who might be interested.

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants