Gordon Phillips and Colleagues Receive NSF Grant to Build New Business Open Knowledge Network

The grant will fund a project linking financial, business, and government information for ease of analysis for entrepreneurs, larger businesses, and researchers.

The National Science Foundation announced recently that it has awarded one of its Convergence Accelerator grants to Gordon Phillips, the Laurence F. Whittemore Professor of Business Administration at Tuck, and computer science and finance colleagues at USC and the University of Maryland.

The Convergence Accelerator NSF grants seek to leverage multidisciplinary research teams to spur public-private partnerships that will allow Fortune 500 companies to apply Big Data to science and engineering to create technologies that enhance the lives of American workers.

The grant is for an interdisciplinary project that will link financial, business and government information for ease of analysis for entrepreneurs, larger businesses, and researchers. The grant investigators will be joined by outside researchers from IBM and OpenCorporates as well as researchers from the U.S. Patent and Trademark Organization (USPTO) and the Securities and Exchange Commission (SEC).

The goal of the project is to create the Business Open Knowledge Network (BOKN), a resource “for entrepreneurs to fully understand the competitive landscape as they create small businesses, allow regulators to quickly identify issues to help prevent the next financial crisis, and enable researchers to develop and test theories to transform our nation's business practices.” Full details of their proposal can be found here.

The grant recipients, experts in finance and computer science, will develop BOKN by building business and finance-specific computational tools that can make sense of unstructured data on the Internet and combine it with structured datasets typically held by governments and regulatory agencies. “Business expertise will drive these computational tools by defining a concrete ontology of concepts, identifying the key entities of interest, and validating the extracted knowledge and downstream predictions in a series of practical use cases,” they say. This work will expand Phillips’s previous efforts on computational linguistics through the Hoberg-Phillips Data Library.

The total grant award is $1 million. The portion going to Dartmouth will support the salary of a new hire in research data services at Tuck as well as for pay computer services at Dartmouth Research Computing on their high performance computers.

Phillips says: “We are excited about the new opportunities the NSF Convergence Accelerator grant provides. It is a multidisciplinary big data grant that uses computational linguistics to analyze over 575 million web pages and 3 million patents from the USPTO, along with financial records from the SEC. The grant will enable us to provide an open knowledge network that both entrepreneurs and academics can use. This open knowledge network will allow users to identify related firms and their patents through an easy to use interface that is open source available to all. It will also allow regulators to identifying emerging systemic risks. Personally, it will allow me to extend my research into the new areas of natural language processing of large-scale databases.”