{"id":342,"date":"2020-10-02T06:47:03","date_gmt":"2020-10-02T06:47:03","guid":{"rendered":"https:\/\/machine-learning.webcloning.com\/2020\/10\/02\/building-an-end-to-end-intelligent-document-processing-solution-using-aws\/"},"modified":"2020-10-02T06:47:03","modified_gmt":"2020-10-02T06:47:03","slug":"building-an-end-to-end-intelligent-document-processing-solution-using-aws","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2020\/10\/02\/building-an-end-to-end-intelligent-document-processing-solution-using-aws\/","title":{"rendered":"Building an end-to-end intelligent document processing solution using AWS"},"content":{"rendered":"
\n

As organizations grow larger in size, so does the need for having better document processing. In industries such as healthcare, legal, insurance, and banking, the continuous influx of paper-based or PDF documents (like invoices, health charts, and insurance claims) have pushed businesses to consider evolving their document processing capabilities. In such scenarios, businesses and organizations find themselves in a race against time to deploy a sophisticated document analysis pipeline that can handle these documents in an automated and scalable fashion.<\/p>\n

You can use Amazon Textract<\/a> and Amazon Augmented AI<\/a> (Amazon A2I) to process critical documents<\/a> and for your NLP-based entity recognition models with Amazon SageMaker Ground Truth, Amazon Comprehend, and Amazon A2I<\/a>. This post introduces another way to create a retrainable end-to-end document analysis solution with Amazon Textract, Amazon Comprehend<\/a>, and Amazon A2I.<\/p>\n

This solution takes scanned images of physical documents as input and extracts the text using Amazon Textract. It sends the text to be analyzed by a custom entity recognizer trained in Amazon Comprehend. Machine Learning applications such as Amazon Comprehend work really well at scale, and in order to achieve 100% accuracy, you can use human reviewers to review and validate low confidence predictions. Additionally, you can use this human input to improve your underlying machine learning models. This is done by sending the output from Amazon Comprehend to be reviewed by human reviewers using Amazon A2I so that you can feed it back to retrain the models and improve the quality for future iterations. You can also use Amazon A2I to provide human oversight to your machine learning models and randomly send some data for human review to sample the output quality of your custom entity recognizer. This automated pipeline can scale to millions of documents with the help of these services and allow businesses to do more detailed analysis of their documents.<\/p>\n

Solution overview<\/h2>\n

The following diagram illustrates the solution architecture.<\/p>\n

\"\"<\/p>\n

This solution takes images (scanned documents or screenshots or pictures of documents) as input. You can upload these files programmatically or through the AWS Management Console<\/a> into an Amazon Simple Storage Service<\/a> (Amazon S3) bucket in the input<\/code> folder. This action triggers an AWS Lambda<\/a> function, TextractComprehendLambda<\/code>, through event notifications<\/a>.<\/p>\n

The TextractComprehendLambda<\/code> function sends the image to Amazon Textract to extract the text from the image. When it acquires the results, it collates the results and sends the text to the Amazon Comprehend custom entity recognizer. The custom entity recognizer is a pre-trained model that identifies entities in the text that are valuable to your business. This post demonstrates how to do this, in detail, in the following sections.<\/p>\n

The custom entity recognizer stores the results in a separate bucket, which acts as a temporary storage for this data. This bucket has another event notification, which triggers the ComprehendA2ILambda<\/code> function. This Lambda function takes the output from the custom entity recognizer, processes it, and send the results to Amazon A2I by creating a human loop for review and verification.<\/p>\n

Amazon A2I starts the human loop, providing reviewers an interface to double-check and correct the results that may not have been identified in the custom entity recognition process. These reviewers submit their responses through the Amazon A2I worker console. When the human loop is complete, Amazon A2I sends an Amazon CloudWatch<\/a> event, which triggers the HumanReviewCompleted<\/code> Lambda.<\/p>\n

The HumanReviewCompleted<\/code> function checks if the human reviewers have added any more annotations (because they found more custom entities). If the human reviewers found something that the custom entity recognizer missed, the function creates a new file called updated_entity_list.txt<\/code>. This file contains all the entities that weren\u2019t present in the previous training dataset.<\/p>\n

At the end of each day, a CloudWatch alarm triggers the NewEntityCheck<\/code> function. This function compares the entity_list.txt<\/code> file and the updated_entity_list.txt<\/code> file to check if any new entities were added in the last day. If so, it starts a new Amazon Comprehend custom entity recognizer training job and enables the CloudWatch time-based event trigger that triggers the CERTrainingCompleteCheck<\/code> function every 15 minutes.<\/p>\n

The CERTrainingCompleteCheck<\/code> function checks if the Amazon Comprehend custom entity recognizer has finished training. If so, the function adds the entries from updated_entity_list.txt<\/code> to entity_list.txt<\/code> so it doesn\u2019t train the model again, unless even more entities are found by the human reviewers. It also disables its own CloudWatch time-based event trigger, because it doesn\u2019t need to check the training process until it starts again. The next invocation of the TextractComprehend<\/code> function uses the new custom entity recognizer, which has learned from the previous reviews of the humans.<\/p>\n

All these Lambda functions use AWS Systems Manager Parameter Store<\/a> for sharing, retaining, and updating the various variables, like which custom entity recognizer is the current one and where all the data is stored.<\/p>\n

We demonstrate this solution in the us-east-1<\/code> Region but, you can run it in any compatible Region. For more information about availability of services in your Region, see the AWS Region Table<\/a>.<\/p>\n

Prerequisites<\/h2>\n

This post requires that you have an AWS account<\/a> with appropriate AWS Identity and Access Management<\/a> (IAM) permissions to launch the AWS CloudFormation<\/a> template.<\/p>\n

Deploying your solution<\/h2>\n

To deploy your solution, you complete the following high-level steps:<\/p>\n

    \n
  1. Create an S3 bucket.<\/li>\n
  2. Create a custom entity recognizer.<\/li>\n
  3. Create a human review workflow.<\/li>\n
  4. Deploy the CloudFormation stack.<\/li>\n<\/ol>\n

    Creating an S3 bucket<\/h3>\n

    You first create the main bucket for this post. You use it to receive the input (the original scans of documents), and store the outputs for each step of the analysis. The Lambda functions pick up the results at the end of each state and collate them for further use and record-keeping. For instructions on creating a bucket, see Create a Bucket<\/a>.<\/p>\n

    Capture the name of the S3 bucket and save it to use later in this walkthrough. We refer this bucket as <primary_bucket><\/em> <\/span>in this post. Replace this with the name of your actual bucket as you follow along.<\/p>\n

    Creating a custom entity recognizer<\/h3>\n

    Amazon Comprehend allows you to bring your own training data, and train custom entity recognition models to customize the entity recognition process to your business-specific use cases. You can do this without having to write any code or have any in-house machine learning (ML) expertise. For this post, we provide a training dataset and document image, but you can use your own datasets when customizing Amazon Comprehend to suit your use case.<\/p>\n

      \n
    1. Download the training dataset<\/a>.<\/li>\n
    2. Locate the bucket you created on the Amazon S3 console.<\/li>\n<\/ol>\n

      For this post, we use the bucket textract-comprehend-a2i-data<\/code>, but you should use the name that you used for <primary_bucket><\/em><\/span>.<\/p>\n

      \"\"<\/p>\n

        \n
      1. Open the bucket and choose Create folder<\/strong>.<\/li>\n
      2. For name, enter comprehend_data<\/code>.<\/li>\n<\/ol>\n

        \"\"<\/p>\n

          \n
        1. Uncompress the file you downloaded earlier and upload the files to the comprehend_data<\/code> folder.<\/li>\n<\/ol>\n

          \"\"<\/p>\n

            \n
          1. On the Amazon Comprehend console, click on\u00a0Launch Amazon Comprehend<\/strong>.<\/li>\n<\/ol>\n

            \"\"<\/p>\n

              \n
            1. Under Customization, <\/strong>choose Custom entity recognition<\/strong>.<\/li>\n<\/ol>\n

              \"\"<\/p>\n

                \n
              1. Choose Train Recognizer<\/strong> to open the entity recognizer training page.<\/li>\n<\/ol>\n

                \"\"<\/p>\n

                  \n
                1. For Recognizer name<\/strong>, enter a name.<\/li>\n<\/ol>\n

                  The name that you choose appears in the console hereafter, so something human readable and easily identifiable is ideal.<\/p>\n

                    \n
                  1. For Custom entity types<\/strong>, enter your custom entity type (for this post, we enter DEVICE<\/code>).<\/li>\n<\/ol>\n

                    At the time of this writing, you can have up to 25 entity types per custom entity recognizer in Amazon Comprehend.<\/p>\n

                    \"\"<\/p>\n

                      \n
                    1. In the Training data<\/strong> section, select Using entity list and training docs<\/strong>.<\/li>\n
                    2. Add the paths to entity_list.csv<\/code> and raw_txt.csv<\/code> for your <primary_bucket><\/em><\/span>.<\/li>\n<\/ol>\n

                      \"\"<\/p>\n

                        \n
                      1. In the IAM role<\/strong> section, select Create a new role<\/strong>.<\/li>\n
                      2. For Name suffix<\/strong>, enter a suffix you can identify later (for this post, we enter TDA<\/code>).<\/li>\n
                      3. Leave the remaining settings as default and choose Train<\/strong>.<\/li>\n<\/ol>\n

                        \"\"<\/p>\n

                          \n
                        1. When the training is complete, choose your recognizer and copy the ARN for your custom entity recognizer for future use.<\/li>\n<\/ol>\n

                          \"\"<\/p>\n

                          Creating a human review workflow<\/h3>\n

                          To create a human review workflow, you need to have three things ready:<\/p>\n