AWS is redefining how companies process documents in a digital world
Think about the last time you opened a bank account, applied for insurance, or refinanced your home. It was probably done on paper. The number of documents in a mortgage packet alone is over 100 pages long. What do you do with all that paper? For many companies across a variety of industries, including financial…
Think about the last time you opened a bank account, applied for insurance, or refinanced your home. It was probably done on paper. The number of documents in a mortgage packet alone is over 100 pages long. What do you do with all that paper? For many companies across a variety of industries, including financial services, healthcare, and manufacturing, processing these documents is painstaking. It’s manual, slow, expensive, and error-prone, and data is often spread across disparate sources. As a result, creating and managing a document processing pipeline remains a challenge for many companies.
According to Ritu Jyoti from IDC, “Supporting document processing requires an AI-native platform that helps improve accuracy, performance, agility and flexibility while supporting a broad set of document types. Artificial Intelligence (AI), can help streamline document automation providing better business outcomes, improved ROI, and reduce manual efforts.”[1]
Today, AWS has launched a solution to help organizations extract insights and automate processing documents of different formats (PDF, Word, raw text) and layouts (bullets, lists) using Amazon Comprehend. This new launch combines the power of natural language processing (NLP) and Optical Character Recognition (OCR) to help reduce the amount of preprocessing or post-processing required to process documents. You can now use custom named entity recognition (NER) on more document types without needing to convert your files to raw text.
AWS has been innovating in the intelligent document processing (IDP) space for years to convert data in documents into usable information for document-centric processes. AWS launched AI services like Amazon Textract, Amazon Comprehend, and others to help with the automation of extracting insights from documents. Since the launch of those services, improvements in accuracy and speed have been tenfold. These services offer new APIs like specialized support for invoices and receipts, handwriting and language support, plus improvements in latency.
Customer success stories
With AWS, customers like Black Knight, Liberty Mutual, and Broadridge Financial Solutions have been able to process millions of documents that come through their pipeline using AWS AI.
Liberty Mutual
Liberty Mutual, a global Fortune 100 property and casualty insurer, offers a wide range of insurance products and services, including personal automobile, homeowners, and many more. With AWS AI, they can process insurance claims at a much higher speed and with greater accuracy.
“At Liberty Mutual, we receive thousands of pieces of information from our customers every day, which our teams then manually input into our business systems before the information can be used by groups within the organization,” says Gillian Armstrong, Solutions Engineer with Liberty Information Technology. “By leveraging Amazon Comprehend, Liberty Mutual has become much more automated, extracting the relevant data from the documents to use in downstream applications, while reducing costs and removing the responsibility from our team of manually entering the data.”
Broadridge Financial Solutions
Broadridge Financial Solutions uses Amazon Comprehend on several SEC filing documents. This ability to extract relevant data from the proxy documents provides insights to inform voting decisions and help institutional investors perform smarter governance. The ability to automate this processing using AWS technology increases scalability and expands the number of data points available to customers, delivering deeper insights, something that couldn’t be done without the use of AI.
“Using AWS intelligent document processing will help us reduce manual process delivery time of up to 7 days,” says said Saumin Patel, VP of Enterprise Architecture at Broadridge Financial Solutions. “Our goal working with AWS was to achieve 70% accuracy on 35 different document elements, which was delivered working with AWS using a combination of OCR and NLP to reduce manual efforts.”
Black Knight
Black Knight, an award-winning provider of software, data, and analytics to the mortgage, real estate, and capital markets verticals uses Amazon Textract to automate its document pipeline. AWS technologies have enabled underwriters to process documents to review results, adjust analyses and request additional documents and information only when necessary.
“Using AWS AI services like Amazon Textract has provided us a way to further automate the underwriting process for our clients, reducing their manual reviews of documents by up to 4 hours. This saves their employees’ time all while improving the mortgage experience to their customers,” said Rich Gagliano, President, Black Knight Origination Technologies.
Partners
AWS APN Partners help enable you to automate your document pipeline and get you to market quicker.
“By leveraging Amazon Machine Learning, our clients are able to refocus their attention on high value business strategies and goals. For example, a global pharmaceutical client of ours leveraged Amazon Textract to improve case entry efficiency, saving them over 9,000 hours a year,” said Evan Scharfer, a Senior Delivery Principal at Slalom.
“As an AWS EU Strategic partner, Kainos are very proud to deliver projects powered by AWS AI which are achieving demonstrable savings for our customers by reducing costs and job processing times over millions of documents and forms. We have proved virtually perfect results with modern documents and surprisingly high levels of accuracy with old, handwritten documents dating back more than 200 years. Our customers have been truly delighted with the results and associated savings, including gaining new insights and enabling faster response times to their customers. AWS continues to prove that it is state of the art in AI” said Laura McKeague, AWS Partner Relationship Manager at Kainos.
Conclusion
The information locked within documents is important to business operations and by using AI, you can now automate the process while reducing manual efforts and improving productivity, which delivers answers to customers faster.
AWS continues to deliver innovation in this space, and today’s announcement of Amazon Comprehend’s integration with Amazon Textract to process custom entities from more document types is just another example of how AWS enables you to extract valuable insights from documents.
[1] IDC Survey Spotlight, What Is the Landscape of the Emerging Document Artificial Intelligence Market?, Doc # US47701421, July 2021
About the Author
Andrea Morton-Youmans is a Product Marketing Manager on the AI Services team at AWS. Over the past 10 years she has worked in the technology and telecommunications industries, focused on developer storytelling and marketing campaigns. In her spare time, she enjoys heading to the lake with her husband and Aussie dog Oakley, tasting wine and enjoying a movie from time to time.
Create your fashion assistant application using Amazon Titan models and Amazon Bedrock Agents
In this post, we implement a fashion assistant agent using Amazon Bedrock Agents and the Amazon Titan family models. The fashion assistant provides a personalized, multimodal conversational experience. Source
In this post, we implement a fashion assistant agent using Amazon Bedrock Agents and the Amazon Titan family models. The fashion assistant provides a personalized, multimodal conversational experience.
Implement model-independent safety measures with Amazon Bedrock Guardrails
In this post, we discuss how you can use the ApplyGuardrail API in common generative AI architectures such as third-party or self-hosted large language models (LLMs), or in a self-managed Retrieval Augmented Generation (RAG) architecture. Source
In this post, we discuss how you can use the ApplyGuardrail API in common generative AI architectures such as third-party or self-hosted large language models (LLMs), or in a self-managed Retrieval Augmented Generation (RAG) architecture.
Visier’s data science team boosts their model output 10 times by migrating to Amazon SageMaker
In this post, we learn how Visier was able to boost their model output by 10 times, accelerate innovation cycles, and unlock new opportunities using Amazon SageMaker. Source
In this post, we learn how Visier was able to boost their model output by 10 times, accelerate innovation cycles, and unlock new opportunities using Amazon SageMaker.