Multidimensional recursive learning process and system used to discover complex dyadic or multiple counterparty relationships

US10019681B2 · kind B2 · utility

Assignee

• THE DUN & BRADSTREET CORPORATION · US

Inventors

• Anthony J. Scriffignano · West Caldwell, US
• David A. Spingarn · Dover, US
• Barry Rizzolo · Allentown, US
• Robin DAVIES · Macungie, US
• Michael Robert Young · Davidson, US
• Laurie Shimer · Stewartsville, US
• John Nicodemo · Bethlehem, US

Key dates

• Filing date: Dec 23, 2014
• Grant date: Jul 10, 2018
• Expiry date: Apr 8, 2036

Classification

• Technology area (CPC G): Physics
• CPC primary: G06Q10/067
• WIPO field: IT methods for management
• WIPO sector: Electrical engineering

Abstract

A multidimensional recursive and self-perfecting process used to discover dyadic or multi-counterparty relationships between parties, the process comprising: (a) collecting information from a plurality of data sources; (b) discovering dyadic or multi-counterparty relationships between the parties from the collected information; (c) clustering the parties to infer the dyadic or multi-counterparty relationships between the parties based on common or partially intersecting attributes between the parties, thereby forming clustered parties; (d) evaluating the clustered parties for business linkage potential by integrating the collected information and contextually assessing indicia from the data sources to detect and measure consistency and inconsistency for a given party or dyadic or multi-counterparty relationship; (e) positing and evaluating relationship type and role said party plays in each relationship; and (f) assessing the confidence level regarding the likelihood that the dyadic or multi-counterparty relationship exists between the parties.