Scrappy Number Two - Automating Loan Approval Processes

[#1]

Team Name	Scrappy Number 2
Track	Banking
Use Case	Automation of our Loan Approval and Underwriting Processes. We want to leverage OCR to scrape text from various underwriting documents, use data quality and match codes for matching, improve banking transaction parsing and improve our time to fund.
Technology	Open Source OCR SAS Visual Text Analytics SAS VDMML SAS Intelligent Decisioning HTTP,XML,JSON packages
Region	Toronto, Ontario, Canada
Team Members	Elmar Taghizade - Risk and Analytics Dorrian Khouri - Real Estate and PM Crysi Popowich - Channel Enablement Alicia Liberty - Central Loan Approval Guilherme Ferreira - Channel Enablement and Insights Namrata Ail - Channel Enablement Surya Surya - Risk and Analytics Katelyn Paterson - Central Loan Approval John Jonkman - Central Loan Approval
Is your team interested in participating in an interview?	Yes
Optional: Expand on your technology expertise	SAS 9.x - 10 Years SAS 3.5 - 3 Years Python - Varied

Pitch Video:

Jury Video:

[jbilenas]

Hard to make loan approval process that would be approved by models very difficult. Also the government requests models used and developed.  Internal validation teams also run validations but have a limited time as well.