Machine Learning Based Document Extraction to Automate Workflows

Company Profile

Leading Fortune 500 Logistics and Transportation services company. They are a provider of truckload, intermodal and logistics services. 

Business Situation

The company receives hundreds of documents during the normal course of business.  These forms are either request for quotes from customers or shipment orders. These forms do not have a consistent layout and each customer may have their own specific format. Today these forms are manually triaged and data entered into their backend systems. This manual process is error-prone and introduces delays. They identified a need to operate more efficiently and increase accuracy by automatically extracting data from these documents.

Google Cloud Implementation

SpringML team was tasked to help with the extraction of intent and entities (e.g. source and destination addresses) from email bodies and PDF/Tiff attachments. Various GCP products such as Dialogflow and Machine Learning API’s were used to build a solution.  There are two primary use cases we worked on – email body extraction and pdf document extraction. To address these challenging use cases, the company partnered with SpringML to pursue a phased approach to prototype GCP’s ML capabilities. SpringML worked on an initial phase to prove the technology.

Thought Leadership

Check out our recent blogs and videos on best practices and implementation approaches.