Skip to content

Mizuho automates data processing of hand-written docs with AI, OCR & RPA

Japan’s Mizuho Financial Group has teamed up with six regional financial institutions to develop an AI-based tool for automating data processing of hand-written documents and non-standard forms.

The internal PoC proved 80% decrease in manual data-entry for direct debit request, across 6 regional financial institutions.

AOR – a combination of of AI (Artificial Intelligence), OCR (optical character recognition) and RPA (Robotic Process Automation) – automates and streamlines the manual data-entry heavy process of dealing with hand-written documents.

In addition to utilising the AOR Solution to substantially improve operational efficiency within Mizuho, the plan is to commercialise it. New company will be established to provide AOR solution to another financial institutions across industries

Mizuho says the project boosts cooperation with regional financial institutions and improve the productivity of FIs overall. It plans to introduce The AOR within its operations sometime this year.

The FS giant has also revealed that it has made an investment in Cogent Labs, an OCR specialist it called in to help on the tool.

The AI Part was supported by Ghelia. Ghelia’s strength lies in its deep learning technology and AI platform development. Mizuho conducted the AOR solution testing jointly with Ghelia.


Published inFintech

Be First to Comment

Leave a Reply