Amazon Comprehend (NLP)

This component analyzes text and extracts detected languages, entities, key phrases and sentiments. This component uses the Amazon Comprehend.

How to Add the Amazon Comprehend NLP Component

  1. Navigate to the AutoClassifier Pipelines component page.
  2. Click New Component and select Amazon Comprehend NLP from the component list:
  3. Name your new Amazon Comprehend NLP component and click Add
  4. Click Apply to save your changes.

  5. Ensure your new Amazon Comprehend NLP component is placed in the list of existing pipeline stages.

How to Configure the Amazon Comprehend NLP Component

To configure your Amazon Comprehend NLP component, select it from the components list and complete the following fields in the Configuration section:

Credentials can be provided one of two ways:

  • Enter AWS Credentials file and AWS Region

  • Directly provide API Key and API secret

  1. If you check Use Credentials file to use a credentials file, specify the following:
    1. Credentials file location: Enter the location of AWS Credential file. For example, C:\Users\Luca\Desktop\credentials.txt

      Example AWS Credentials File

      Copy
      [{profilename}]
      aws_access_key_id = {accessKey}
      aws_secret_access_key = {secretKey}
    2. Credentials Profile Name: Enter AWS Profile Name for the Credentials File.
  2. If you did not check Use Credentials file to use a credentials file, specify the following:
    1. Api Key: Setup your Amazon Comprehend instance, obtain an API key, and enter your key into this field.
    2. Api Secret: Enter the API Secret of your Amazon Comprehend instance.
  3. Amazon Web Service Region: Select the region of your Amazon Web Service from the drop-down list. The supported Regions for Amazon Comprehend are documented here.
  4. Input Property: Enter the property that is configured for entity extraction. The default value for this field is 'body'.
  5. Extract Languages: Enable this checkbox to extract and output the languages that are detected in the current document.
  6. Language score threshold: Specify a value between 0 and 1 that represents the minimum confidence score accepted for a detected language.
  7. Use detected language: Enable this checkbox to use the detected language with the highest confidence to analyze the text for entities, key phrases, or sentiments.
  8. Default Language: From the drop-down list, select the predefined language that is used to analyze the text if no other detected language is found.
  9. Extract entities: Enable this checkbox to detect entities in the input text.
  10. Entities score threshold: Enter a value between 0 and 1.
  11. Entities “No. of Matches” threshold: Specify the minimum number of occurrences for a given entity to be included in the output results.
  12. Maximum Entities: Specify the maximum number of entities to return. Results will be returned by occurrence.
  13. Extract key phrases: Enable this checkbox to detect key phrases in the input text.
  14. Key Phrase score threshold: Enter a value between 0 and 1. 
  15. Key Phrase “No. of Matches” threshold: Specify the minimum number of occurrences for a given entity to be included in the output results.
  16. Maximum Key Phrases: Specify the maximum number of key phrases to return. Results will be returned by occurrence.
  17. Extract sentiments: Enable this checkbox to analyze the input text to determine whether it contains negative or a positive content. 
  18. Send raw response as metadata: Enable this checkbox to attach the JSON file response represented all analyzed results.

Output Properties

Property

Type

AmazonExtractedLanguages

Text – Multi

AmazonExtractedEntities

Text – Multi
AmazonExtractedPhrases Text – Multi
AmazonExtractedSentiment Text
AmazonRawResponse Text
ComprehendSerializedEntitiesJson