An entrepreneurial team including the principal investigator Mequanint Moges, engineering technology department assistant chair and instructional associate professor; Russ Gundrum, engineering technology lecturer and industry lead; Salah Ahmad, engineering technology lecturer and entrepreneurial lead; and Joanna Russian, PhD student and co-entrepreneurial lead, has been awarded a $50,000 National Science Foundation (NSF) Innovation Corps I-Corps ™ grant. The grant will support their research, "An Automatic Data Capture System Using Machine Learning: Enhancing Accuracy and Productivity."
Businesses like banking and mortgage, manufacturing, healthcare, and government that rely heavily on paper-intensive processes, data entry, and document production are often plagued with inaccuracies that impact cost, efficiency and productivity. "Integrating robotic process automation into the data entry business processes in a structured way accurately extracts and processes data much faster at a relatively low cost point," said Salah Ahmad.
The proposed system eliminates traditional methods of data extraction and document classification by using a self-learning software program to gather information from any scanned document. An innovative framework that explores the application of convolutional neural networks (CNNs) to data capture and processing is a proprietary non-traditional approach that identifies and extracts information from scanned documents, making it possible to fully automate paper-intensive business processes in less time. Similar technologies such as optical character recognition (OCR), are not capable of the required level of accuracy.
With the goal of transitioning their work into a commercial process or product, The I-Corps team will conduct 100 interviews and business model presentations with potential customers.