The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
Sep. 12, 2006
Filed:
Mar. 31, 1999
Quan G. Cung, Austin, TX (US);
Harry Roger Kolar, Scottsdale, AZ (US);
Kevin Eric Norsworthy, Austin, TX (US);
Julio Ortega, The Colony, TX (US);
Frederick J. Scheibl, Austin, TX (US);
Vasken Torossian, Round Rock, TX (US);
Ben Peter Yuhas, Baltimore, MD (US);
Quan G. Cung, Austin, TX (US);
Harry Roger Kolar, Scottsdale, AZ (US);
Kevin Eric Norsworthy, Austin, TX (US);
Julio Ortega, The Colony, TX (US);
Frederick J. Scheibl, Austin, TX (US);
Vasken Torossian, Round Rock, TX (US);
Ben Peter Yuhas, Baltimore, MD (US);
International Business Machines Corporation, Armonk, NY (US);
Abstract
Attributes of a data set to be employed in generating a predictive model are analyzed based on entropy, chi-square, or similar statistical measure. A target group of samples exhibiting one or more desired attributes is identified, then remaining attribute values for the target group are compared to corresponding attribute values for the whole sample population. A subset of all available attributes is then selected from those attributes which exhibit, when comparing attribute values of target group samples to attribute values for the whole sample population, the greatest relative difference or divergence. This subset is employed to generate the predictive model. Efficiency in generating the predictive model and the accuracy of the resulting predictive model is improved, since fewer attributes are employed and less computational resources are required.