0

Using the NLU Analyze for Entities and Keywords, the response returns the following Sentiment item score: Sentiment score for the concept ranging from -1 to 1. Negative scores indicate negative sentiment, and positive scores indicate positive sentiment.

Does this score represent the magnitude of the positive or negative sentiment (i.e. -.9952 is very negative while -.1301 is sort of negative)? OR Does this score represent the confidence of the positive or negative sentiment (i.e. -.9952 is definitely negative while -.1301 only probably negative)

https://www.ibm.com/watson/developercloud/natural-language-understanding/api/v1/#entities https://www.ibm.com/watson/developercloud/natural-language-understanding/api/v1/#keywords https://www.ibm.com/watson/developercloud/natural-language-understanding/api/v1/#sentiment

data_henrik
  • 16,724
  • 2
  • 28
  • 49
Kacey Tovornik
  • 1
  • 1

1 Answers1

0

The score represents the magnitude of the sentiment. In addition to the links you referenced, I found this Redbook on the IBM Watson NLU (Natural Language Understanding) service quite interesting. It is linked off the additional resources in the NLU docs.

data_henrik
  • 16,724
  • 2
  • 28
  • 49