“Unstructured data” refers to data that are often text-based (as opposed to numeric), and have no data model. Yet, the information contained in this type of data can lead to significant business and financial intelligence.
The applied research question we're addressing is whether empirical data and machine-learning tools can help us understand financial risk profiles in an industry-specific context?
Since 2007, we developed and patented an algorithmic approach to uncover risk and opportunity in private investments. Using unstructured data from social networks and other publicly available resources, and the context of industry value chains, our KeyStone Compact® algorithms quantitatively profile the business model risks of startups, SMEs and corporate lines of business.
The analysis is typically conducted in two stages:
Drawing from research on strategy, marketing models, and business valuation tools, the KeyStone Compact® risk model analyzes the company through two lenses: Industry value chain position; and Investment Grade (tied to market access opportunity) .
These tools are commercialized by Corymbus Asset Management and one of its channel partners, the Global CleanTech Cluster Association. To date, 1000's of companies have been analyzed and positioned, and over $500M. of debt and equity capital invested in selected companies.
Nicole Bette, MSc, Biomedical Engineering; Kavya Vayyasi, MSc, Electrical Engineering and Computer Science; Nic Miller, BSE, Civil and Environmental Engineering; Sven Adriaens, BBA, Finance, Michigan State University
Alex Mercier, MSc (EWRE), IT and Cyber Security management Consultant at Wavestone, UK; Antti Tahvanainen, PhD, Research Institute of the Finnish Economy, Helsinki, Finland; Tim Faley, Professor and Sokoloff Chair of Entrepreneurship, University of the Virgin Islands; David Brophy, Professor, Ross School of Business.