Given a set of labels in text format seperated by commas and text to classify, this node will return a JSON string object which a confidence score attached to each label.

Outputs

This node outputs a table. the table will have two columns, labels and scores.

Example

Given the input:

Text: From where can I manage my billing?

Labels: account, bug fix, feature request

It would return the following JSON string object:

 [{ "labels": "feature request", "scores": 0.022516 },
  { "labels": "bug fix", "scores": 0.0212345 },
  { "labels": "account & billing", "scores": 0.99 }]

Use Cases

This can be used for a variety of use cases, in the above example we are setting up a scenario where a user is interacting with a support team and the text classifier helps us identify which team to forward the user's message to.