detect_targeted_sentiment

detect_targeted_sentiment(**kwargs)

Inspects the input text and returns a sentiment analysis for each entity identified in the text.

For more information about targeted sentiment, see Targeted sentiment.

See also: AWS API Documentation

Request Syntax

response = client.detect_targeted_sentiment(
    Text='string',
    LanguageCode='en'|'es'|'fr'|'de'|'it'|'pt'|'ar'|'hi'|'ja'|'ko'|'zh'|'zh-TW'
)

Parameters

• Text (string) --

  [REQUIRED]

  A UTF-8 text string. The maximum string length is 5 KB.

• LanguageCode (string) --

  [REQUIRED]

  The language of the input documents. Currently, English is the only supported language.

Return type

dict

Returns

Response Syntax

{
    'Entities': [
        {
            'DescriptiveMentionIndex': [
                123,
            ],
            'Mentions': [
                {
                    'Score': ...,
                    'GroupScore': ...,
                    'Text': 'string',
                    'Type': 'PERSON'|'LOCATION'|'ORGANIZATION'|'FACILITY'|'BRAND'|'COMMERCIAL_ITEM'|'MOVIE'|'MUSIC'|'BOOK'|'SOFTWARE'|'GAME'|'PERSONAL_TITLE'|'EVENT'|'DATE'|'QUANTITY'|'ATTRIBUTE'|'OTHER',
                    'MentionSentiment': {
                        'Sentiment': 'POSITIVE'|'NEGATIVE'|'NEUTRAL'|'MIXED',
                        'SentimentScore': {
                            'Positive': ...,
                            'Negative': ...,
                            'Neutral': ...,
                            'Mixed': ...
                        }
                    },
                    'BeginOffset': 123,
                    'EndOffset': 123
                },
            ]
        },
    ]
}

Response Structure

• (dict) --

  • Entities (list) --

    Targeted sentiment analysis for each of the entities identified in the input text.

    • (dict) --

      Information about one of the entities found by targeted sentiment analysis.

      For more information about targeted sentiment, see Targeted sentiment.

      • DescriptiveMentionIndex (list) --

        One or more index into the Mentions array that provides the best name for the entity group.

        • (integer) --

      • Mentions (list) --

        An array of mentions of the entity in the document. The array represents a co-reference group. See Co-reference group for an example.

        • (dict) --

          Information about one mention of an entity. The mention information includes the location of the mention in the text and the sentiment of the mention.

          For more information about targeted sentiment, see Targeted sentiment.

          • Score (float) --

            Model confidence that the entity is relevant. Value range is zero to one, where one is highest confidence.

          • GroupScore (float) --

            The confidence that all the entities mentioned in the group relate to the same entity.

          • Text (string) --

            The text in the document that identifies the entity.

          • Type (string) --

            The type of the entity. Amazon Comprehend supports a variety of entity types.

          • MentionSentiment (dict) --

            Contains the sentiment and sentiment score for the mention.

            • Sentiment (string) --

              The sentiment of the mention.

            • SentimentScore (dict) --

              Describes the level of confidence that Amazon Comprehend has in the accuracy of its detection of sentiments.

              • Positive (float) --

                The level of confidence that Amazon Comprehend has in the accuracy of its detection of the POSITIVE sentiment.

              • Negative (float) --

                The level of confidence that Amazon Comprehend has in the accuracy of its detection of the NEGATIVE sentiment.

              • Neutral (float) --

                The level of confidence that Amazon Comprehend has in the accuracy of its detection of the NEUTRAL sentiment.

              • Mixed (float) --

                The level of confidence that Amazon Comprehend has in the accuracy of its detection of the MIXED sentiment.

          • BeginOffset (integer) --

            The offset into the document text where the mention begins.

          • EndOffset (integer) --

            The offset into the document text where the mention ends.

Exceptions

• Comprehend.Client.exceptions.InvalidRequestException
• Comprehend.Client.exceptions.TextSizeLimitExceededException
• Comprehend.Client.exceptions.UnsupportedLanguageException
• Comprehend.Client.exceptions.InternalServerException