A Machine Learning Approach to Helping Small Businesses Find Pandemic Economic-Impact Relief. Issue 3 (3rd July 2021)

• Record Type: Journal Article
• Authors: Czapski, Michal; Godfrey, Stephen; Derenski, Joshua; Khader, Isaac
• Is Part Of: Chance. Volume 34:Issue 3(2021)
• Page Start: 61
• Page End: 68
• Publication Date: 2021-07-03
• Subjects: Statistics -- Periodicals; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Probablities -- Periodicals; 519.205
• DOI: 10.1080/09332480.2021.1979820
• Languages: English
• ISSNs: 0933-2480

Abstract

Abstract : While the Global Health Organization was able to officially declare the spread of COVID-19 as a global pandemic late in Q1 2020, the most effective responses from both governmental and private organizations were by no means clear. Very little was known about what was then frequently referred to as the novel coronavirus, and medical professionals had few recommendations specific to this disease. Still, what was abundantly clear was stay-at-home and lockdown orders were needed to bend the curve or slow transmission. As customers sheltered in place and businesses closed their doors, the impact on small businesses was expected to be devastating. With so many sources for potential aid from U.S. governments, and private and philanthropic entities available C2CB, aided by SWB, focused on helping small businesses identify relevant aid resources. SWB, consulting with C2CB, built a multistage data pipeline using machine learning techniques to automatically curate a national list of small-business aid programs, presenting users with results to efficiently research and find relevant aid programs. While this project curates business relief grants, it is a proof-of-concept for a no-cost data pipeline using machine learning techniques with automated website relevancy classification.