Method and system for clustering optimization and applications

US6745184B1 · kind B1 · utility

Assignee

• Rosetta Marketing Strategies Group · US

Inventors

• Lawrence Choi · Downey, US
• Christopher B. Kuenne · Princeton, US
• Kurt E. Holstein, II · Princeton, US
• Henry Andrew Cross · Moorestown, US
• George Tang · Saratoga, US
• Chetna Bansal · New York, US
• Jason Whitney · Lee's Summit, US
• Joshua D. Babbitt · Jersey City, US

Key dates

• Filing date: May 31, 2001
• Grant date: Jun 1, 2004
• Expiry date: Aug 9, 2022

Classification

• Technology area (CPC Y): Emerging Cross-Sectional Technologies
• CPC primary: Y10S707/99942
• WIPO field: IT methods for management
• WIPO sector: Electrical engineering

Abstract

A computer-assisted method for placing each of a plurality of observations into one of a predefined set of clusters is disclosed. The disclosed method includes, for each of a plurality of observations, obtaining a data set containing no more than one proxy value for each of a plurality of variables, each variable having a plurality of possible values, and obtaining a set of first dummy variables having a value indicative of an observation's membership or non-membership (i.e., membership status) in a particular cluster. The disclosed method also includes obtaining a set of second dummy variables each having a value indicative of an observation's having a particular response or lack thereof for a given variable, and, for each observation, regressing the set of first dummy variables on the set of second dummy variables to obtain a linear approximation to a plurality of probabilities that an observation with particular provided values to the plurality of possible values belongs to each particular cluster from the predefined set of clusters.