Skip to main content Skip to complementary content

Amazon Comprehend analytics source

Amazon Comprehend is a managed service for text analysis available to customers of the AWS platform.

The following Amazon Comprehend services are supported:

  • Sentiment Analysis

  • Key phrase Extraction

  • Language Detection

  • Entity Recognition

  • Syntax Analysis

  • Personally Identifiable Information (PII) Detection

Each of these services requires a separate connection to be created with the specific parameters applicable to the service in question.

https://aws.amazon.com/comprehend/.

Limitations

  • Amazon Comprehend has several quotas and limitations that can be increased using the AWS Service Quotas and AWS Support Center.

    For example, the document size (UTF-8 characters) is 5000 bytes which means that the limit per app row size in Qlik Sense is 5000 bytes.

    Using the Amazon Comprehend endpoint with data rows larger than 5K bytes will cause an TextSizeLimitExceededException error. Such an error message can for example state: "Input text size exceeds limit. Max length of request text allowed is 5000 bytes while in this request the text size is 5276 bytes"

    https://docs.aws.amazon.com/comprehend/latest/dg/guidelines-and-limits.html.

  • The Amazon Comprehend connector is limited to 1000 rows per request. These are sent to the endpoint service in batches of 25 rows. In scenarios where more rows are required to be processed, use a Loop within the data load script to process more rows in batches.

  • The Amazon Comprehend Personally Identifiable Information (PII) Detection service is limited to 100 rows per request. These are sent to the endpoint service 1 at a time.

  • In a scenario where an application is regularly reloaded, it is best practice to cache the predictions using a QVD file and only send the new rows to the prediction endpoint. This will improve the performance of the Qlik Sense application reload and reduce the load on the Amazon Comprehend endpoint.

  • There is a known limitation at the Amazon Comprehend side, where sending null values, or blanks, to Amazon Comprehend will result in an error returned by Amazon Comprehend.

  • If you are using a relative connection name, and if you decide to move you app from a shared space to another shared space, or if you move your app from a shared space to your private space, then it will take some time for the analytic connection to be updated to reflect the new space location.