What is sentiment polarity and where to find it in Content Analysis API results?

Sentiment analysis is the process of inspecting a given text to determine the author’s attitude to the described topic. A basic task in sentiment analysis is classifying the sentiment polarity of a given text – whether the expressed opinion is positive, negative, or neutral.

Every endpoint of the Content Analysis API provides data on the sentiment polarity of the citation for the target keyword. To discover sentiment value you need to find the "connotation_types" object in the result. It contains 3 types of sentiments related to the keyword citation: "positive", "negative", "neutral", and their count.

You can find the example in the results of all 4 Content Analysis API endpoints. Here we have a part of the Content Analysis – Summary API‘s response:

{
   "connotation_types": {
       "positive": 1506520,
       "negative": 898628,
       "neutral": 1366490
 }
}

Here you can see the exact count of citations with positive, negative, or neutral sentiment polarity.