Sentiment: Calculate score

Use this action to automatically detect positive or negative language or feelings present within an email, document or block of text.

Use Cases

• Analyze inquiries from customers, use this action to flag negative sentiment to prioritize responses and route the inquiry to the most appropriate representative
• Analyze customer comments over time to determine a customer satisfaction score
• Include sentiment and magnitude in a predictive model to categorize or prioritize inquiries

How to configure this action

Sentiment analysis supports the following languages:

Fields for this action

• Text for analysis

  • The text for which you would like to calculate sentiment. Typically, you will reference a field for this.

What will this output?

The output of this action is the sentiment and magnitude score for the analyzed text and the predicted language. For example, the text “clearly positive” has a sentiment score of .8 and magnitude of 3.0. Other examples:

Output fields for this action

• Sentiment

  • Overall emotion of the text on a scale from -1.0 (very negative) to 1.0 (very positive)

• Magnitude

  • How much emotional content is present within the document, which is usually proportional to the length of the document

• Language

  • The language of the text using an ISO 639-1 code, for example en for English or de for German

Tips

• Use task conditions to determine actions that should occur based on sentiment and magnitude scores