BAL is one of the world’s most recognized corporate immigration law firms, and is the Best Lawyers® “Law Firm of the Year” in Immigration Law in America for 2019. The firm is singularly focused on meeting the immigration challenges of corporate clients around the world in ways that make immigration more strategic and enable clients to be more successful. Established in 1980, the firm provides immigration expertise, top-notch information security and leading technology innovation like its Cobalt® digital immigration services platform.
BAL Global receives hundreds of documents during the normal course of business. These forms are manually scanned to PDF and the data on the forms are entered into various systems. The intention of this PoC is to apply automated techniques (e.g. OCR) to classify, separate, sort, and extract data from these documents. They identified a need to operate more efficiently and increase accuracy by automatically classifying, separating, sorting and extracting data from documents scanned to PDF format.
Google Cloud Implementation
SpringML team was tasked to help with building parts of this solution using Google Cloud Platform. The technology stack included Vision API to extract text from the documents and App Engine for orchestration. Several documents like I797, I94 were parsed and ML based techniques were applied to extract key fields like Receipt Number, Receipt Date, Notice Date, etc. The output of this project can be used automate document parsing and field extraction thereby reducing manual effort.
BAL Global was able to automate the immigration document processing and reduce human errors using the implemented approach.
SpringML’s document processing solution can enable us to augment our staff and processes with Machine Learning. The system delivers the capability to enhance our client service offerings through supplementing quality review processes with smart systems, while at the same time reducing paper processing time. This allows our legal teams to take action faster for our clients. This solution is versatile in that it handles several types of documents. It also provides details on data extraction accuracy, which allows us to be confident of the results.
Check out our recent blogs and videos on best practices and implementation approaches.