Arcanum implemented an AI powered invoice extraction workflow to supercharge Core Loyalty’s ability to extract data when processing invoices. The AI technology means there are no templates needed to accurately read a wide range of invoices in order to extract key information with higher confidence and accuracy rate.

AWS Solution Architecture

The invoice extraction workflow utilises AWS Textract, a machine learning service that automatically extracts text, handwriting and data from document sources. Combining queries with the out of the box capabilities of the analyse expense API, we were able to build our own post processors on top of Textract to create a robust, accurate, and generalised invoice extraction service for Core Loyalty. We also utilised Route 53, VPC, Internet Gateway, Load Balancer, NAT gateway, ECS, ECR, EC2.

Implementation

For Core Loyalty, we were able to utilise the powerful queries feature available in Textract to dynamically handle various types of invoice formats. The capabilities of queries meant that we weren’t limited by changing templates and vendor invoice styles due to their advanced combination of NLP and CV utilisation, as well as the functionality to assign multiple queries to a single alias. For example we could search for the invoice number using multiple queries to broaden the search, such as “What is the invoice number” as well as “What is the document number”. Combining queries with the out of the box capabilities of the analyse expense API, we were able to build a robust, accurate, and generalised invoice extraction service for Core Loyalty.